Slashdot Mirror
Mirroring Controllers - What have been Your Experiences?
Today's installment is a lengthy (but hopefully informative) piece on mirroring controllers. Ever had weird problems with a FastTrack TX2000? Are you curious how well HighPoints RocketRAID boards are really supported? Ever wondered which controller gives you the best performance for every dollar spent? In true Slashdot tradition, we're taking the issue and throwing it out to you, the readers. Futurepower(R) is willing to start us off with a wealth of information on his experiences, and I'm hoping a few of you are willing to do the same.
Futurepower(R) asks: "What experiences, both good and bad, have you had with mirroring controllers? Are there manufacturers I haven't found?
For those who are interested but don't have experience, mirroring controllers (RAID 1) provide several advantages:
ECS (EliteGroup) has made 11 motherboards with on-board RAID mirroring controllers. One of them, the P4VXAD, has a Promise controller and costs about $52. What's that about? Employees at Fry's tell me that ECS has the poorest quality control of any of the motherboard manufacturers that they sell. To me, ECS motherboards seem to have surprisingly high quality. However, we have only tested three, and only one extensively.
Silicon Image makes, or made, their 0680 RAID chipset, and Koutech Systems sells the IOFLEX-Pir133 using that chipset. It costs $25 retail. I tried a Koutech card and had a lot of trouble with it, even after updating the BIOS. I talked to an application engineer at Silicon Image and a manager there. I was told that SI bought the technology from another company, and apparently it is no longer supported, doesn't work well, and the company does not intend to put more money into it. I found that the Koutech card corrupts files. The card gives a Stop 07B error when going into the Windows XP Recovery Console, unless the driver is loaded by hand, every time. I lost a lot of time with a product that apparently should not be sold.
My experience with mirroring controllers is mostly with those from Promise Technology. I've been using Promise mirroring controllers since they began making them.
The good about Promise:
The Acard controller supports SuSE, Red Hat, Caldera, and Turbo Linux, it says. But remember, Froogle found only one vendor.
HighPoint says they support Linux: "Linux Red Hat 7.3 & 8.0 (Software RAID Only)". This apparently means the card does not support Linux at all, since Linux has software RAID built in.
So, that's the extent of my knowledge and experience. Can you provide further insight?"
For those who are interested but don't have experience, mirroring controllers (RAID 1) provide several advantages:
- They prevent data loss when a hard drive fails. The other drive in the mirror takes over seamlessly.
- Reading of data is considerably faster since the controller reads the data from the drive that has a head closest to the data.
- You always have a full hard disk backup that you can pull from your system at any time, such as before installing new software.
- You can clone a Windows XP drive using the RAID card, and you will have a bootable copy. This is valuable, since the Windows XP file system cannot copy all of its own files. (Don't flame this; it has been verified many times by Microsoft employees, who often suggest using the third-party tools.)
- Promise Technology's FastTrak TX2000 is available for about $85 delivered.
- HighPoint's RocketRAID 133 costs about $80.
- ACARD Technology's AEC-6880 costs about about $85. Froogle found only one vendor. That's scary.
- Adaptec makes the ATA RAID 1200A, which is available for about $59 before delivery charges.
ECS (EliteGroup) has made 11 motherboards with on-board RAID mirroring controllers. One of them, the P4VXAD, has a Promise controller and costs about $52. What's that about? Employees at Fry's tell me that ECS has the poorest quality control of any of the motherboard manufacturers that they sell. To me, ECS motherboards seem to have surprisingly high quality. However, we have only tested three, and only one extensively.
Silicon Image makes, or made, their 0680 RAID chipset, and Koutech Systems sells the IOFLEX-Pir133 using that chipset. It costs $25 retail. I tried a Koutech card and had a lot of trouble with it, even after updating the BIOS. I talked to an application engineer at Silicon Image and a manager there. I was told that SI bought the technology from another company, and apparently it is no longer supported, doesn't work well, and the company does not intend to put more money into it. I found that the Koutech card corrupts files. The card gives a Stop 07B error when going into the Windows XP Recovery Console, unless the driver is loaded by hand, every time. I lost a lot of time with a product that apparently should not be sold.
My experience with mirroring controllers is mostly with those from Promise Technology. I've been using Promise mirroring controllers since they began making them.
The good about Promise:
- Promise has been selling RAID 1 controllers a long time. They work.
- Promise controllers can clone a hard drive quickly.
- Now some motherboards have Promise mirroring chipsets. You can get the entire motherboard with the Promise controller on the board, for maybe $50 more than the Promise controller card alone.
- Linux drivers are available. The web site says, "Windows XP/2000/NT4/Me/9x; Novell NetWare 4.1x/5.x; RedHat Linux 7.0/7.1/7.2; TurboLinux Server 6.5; TurboLinux Workstation 7; SuSE Linux 7.2; OpenLinux 3.1" Does that mean that later versions of Linux can't use this card, or did Promise forget to update the brochure PDF file?
- Promise controllers work fine under DOS, but there is no error reporting if for some reason the mirror breaks.
- Promise mirroring controllers have a software feature called "sychronization". I've asked many times over the years why it is necessary, since mirrored hard drives should be synchronized 100% of the time. I've never gotten an answer. Recently I've been told by Promise technical support people not to use synchronization, since it has caused problems. It sounds like some technical problem is being hidden.
- Promise does not support their oldest mirroring controllers under Windows XP. This is a problem since there are many business computers that are used for data entry. A Pentium II is as fast as is necessary. Windows 98 is stable with only one program running. Now those computers need to be converted to Windows XP, since Microsoft has declared that its operating systems have a curious quality: They die. (According to Microsoft, it doesn't matter that at least 100,000,000 people are using Windows 98 worldwide, it came to the end of its life on "30-Jun-2003".)
So, it is necessary to buy another controller for old data entry systems. Notice that Microsoft and Promise could decide to play this game again, and I would like to avoid the second round of buying and installing even another controller. I'd like to find a company that continues to support its products.
The speed of computers used for data entry does not matter, but the security of the data does. Hard drive failures are becoming rare, but a hard drive failure can cause a lot of problems on a data entry computer, so mirroring is required. - Some Promise controllers, especially those on motherboards, take a long time to boot. Dots crawl across the screen even if no drives are connected to the controller. Is keeping the Promise name on the screen a time-wasting sales message from Promise? Recently Promise released a BIOS upgrade for some of its cards that reduced the dot-crawling time. However, there is apparently no upgrade for Promise controllers on motherboards.
- About 2 months after I reported problems, Executive Software said they found a bug in their Diskeeper defragmentation software that might cause data corruption when used with Promise controllers. They said everyone using Diskeeper should upgrade to the new, free, minor version. I've seen no problems since then.
- Promise Technology's sales literature can be disgusting. This is the second sentence in Promise's description of the FastTrak TX2000: "The FastTrak TX2000 ATA RAID card supports Ultra ATA/133 drives to rock workstations and boost small (or large) office servers like never before." To me, this is obviously written by someone who knows nothing about the product and doesn't care.
- I find the abundant use of PDF files and unnecessary JavaScript on Promise Technology's web site annoying.
The Acard controller supports SuSE, Red Hat, Caldera, and Turbo Linux, it says. But remember, Froogle found only one vendor.
HighPoint says they support Linux: "Linux Red Hat 7.3 & 8.0 (Software RAID Only)". This apparently means the card does not support Linux at all, since Linux has software RAID built in.
So, that's the extent of my knowledge and experience. Can you provide further insight?"
I just want to say that it's absolutely impossible to read that much italic. My eyes are bleeding.
look
That's all I'm sayin...
Ever had weird problems with a FastTrack TX2000?
No, but I have had weird problems with a T-1000 a few times.
Refuse to make a statement in your sig!
Can you spell 3ware?
They do more than just mirroring, and aren't cheap, but if you wants the quality, you gots ta pay the piper.
Drivers are in the Linux kernel, and have been for some time. ATA or S-ATA versions available.
Wrong... no, 'Wrong' assumption.
Short version: hardware RAID, not software trickery. The drivers are in the normal Linus kernel, so you don't have to go tracking down some external hackjob crap. You plug in IDE disks and it looks like SCSI to the host. Buy the right kind of cage and you can even do hotswap.
Hey, when you work with machines that are all hotswappable RAID and come home to machines that are pathetic single-disk IDE junk, you start wanting something more.
In true Slashdot tradition, we're taking the issue and throwing it out to you
and in true Slashdot tradition, the result of this will likely be in-soviet-russia, 123-profit, sco-sucks or imagine-a-beowulf-cluster-of-these.
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
Both cards are good quality.
One additional thing that should be mentioned is not only the Raid 1, but the fact that (at least the adaptec) they also support Raid 0 and 0+1. This means that with 4 drives (2 on each IDE chain) you can stripe the data between two drives AND have it mirrored. Fast as hell reads and faster writes.
I don't know if the promise card has it, but it should.
I personally use a promise IDE controller at home.. it is rock solid. Just make sure that you do do their firmware updates. It does make a difference.
As far as I see it.. 1 vendor = bad, promise = 'we make controller cards... we will expand into raid cards', adaptec = 'ohh...people want cheaper raid, time to go IDE'. I have had a adaptec 1200A go south on me once, but luckily I hadn't finished setting up the system.
RoundTop
Heh, while waiting to post again, I had a crisis of confidence. I wondered if SUCCEED was spelt correctly. As I had Google in another window, I put SUCEED into it, and clicked for a search.
Naturally, it came back "Did you mean 'Succeed'", but the first link returned by Google, on a search for a misspelling of the word succeed is:
Well looky here
Now THATS ironic, Alanis.
My thoughts on the Promise FasTrack controllers:
We use 'em at work. On the Windows 2000 side, they come with decent management software and for the most part, are relatively reliable.
However, they are far from perfect. I've had several W2K servers blue screen when doing a hot-swap. Joy.
The FreeBSD drivers are bloody stable as hell. No complaints.
The Linux drivers provided by Promise are, IMHO, a POS. Pain to compile. No management software. Diagnositics are limited. As a result, I'd go with a different IDE controller card if you want it for Linux. YMMV.
Here are some mirrors.
The Adaptec ATA RAID 1200A uses the HPT370 chipset so its basically just a HighPoint card.
Promise is an evil company what doesn't provide any type of support for it's users... I have a $1000 RAID 5 card here, that doesn't even have Windows XP drivers... 3ware... proper Linux drivers, excelent support... once again 3WARE...
Why no mention of the 3ware 7000-2?
Easily beats most if not all of these cards, especially since it is a hardware card, unlike some of those software cards mentioned.
Do you remember Super Dave Osborn?
How does running Linux make you immune to physical hard drive failures? The hard drive doesn't care whether it has a windows or linux filesystem on it when it dies.
Not considerably. Faster, yes, but nothing like Raid 0.
Raid 1 isn't striped, a la, your data blocks alternate sequentially between drive 0 and drive 1. Raid 1 basically issues a read request to both drives and uses the data returned from whichever drive responds first. As you said, the one whose head was closer to the data.
Contrast that with Raid 0 (or Raid 10), where blocks are read in parallel (block N from drive A, block N+1 from drive B, etc.) That parallelism leads to close to 2x read performance (on sequential reads). Versus maybe 1.1x performance improvement on Raid 1.
I have used several Promise FastTrak RAID controllers and have had varied results, but all in the "crappy" range. Their drivers are proprietary, so you have to stick with the pre-compiled kernel module. This means you also have to stick with a pre-compiled distro kernel so the symbols match. Promise has always been several kernel versions behind, so if there is a kernel security upgrade, you have two choices: 1. patch the kernel and break the Promise card or 2. Leave a vulnerable computer up and running. You can generally force a promise module into a non-matching kernel, but I've always been hesitant to do that.
I have switched every Promise installation to 3ware cards because of this. They are open source drivers, very current, and perform well. Their tools run as a web daemon under Linux so you can check status/reconfig on the fly. Really amazing Linux support, and a reasonable price... (and no, I'm not affiliated...:) )
"Just another damned fool idealistic crusader..."
Okay, I'll feed the troll...
Does your hardware never fail? Does your filesystem never get corrupted? Power never goes out? Do you administer things in some sort of fantasy world of perfection?
Thats pretty impressive.
Mylex controllers work very well for higher end SCSI workstations and servers. I've used these in a few systems and they are very easy to setup in both linux and windows, offer great performance, and well supported. A little on the expensive side though..
I've used the Adaptec card before, and I learned the hard way that you need to be aware of heat. Two modern high-speed IDE drives generate a large amount of heat, and if you mount the two directly next to one another in the case you may well burn one out, as I did.
The dead drive may have had some other defect, but I doubt it as the thermal alarm on the MB would sound after a few hours of usage. Even after replacing the drive (and being sure the two drives were mounted far enough apart to allow airflow), the machine still occasionally comes up with blank disks, requiring an OS reinstall.
Just something to keep in mind when you're building your system--YMMV.
You are an idiot!
I am the root bridge.
usually (afaik/iirc/if_i_got_a_clue) the cards that are cheap use the cpu on motherboard for the raid functionality, and as far as i'm concerned i find their 'only' use to be cheap(ish) way to add more ata drives to a system.
get a real card or you might just as well use software raid, the good models don't come with 64mb of mem just for kicks either, and the good models have linux support too (adaptec 2400a, if that was the model i installed once, i think it is, even came with managing software that was a bootable linux cd). sure it's expensive but it's REAL, not just an ata interface and programs.
world was created 5 seconds before this post as it is.
Ever had weird problems with a FastTrack TX2000?
Does having a FastTrack 100 TX2 die completely causing me to lose everything on a RAID 0, I know I was asking for it anyway, constitute a problem?
Data loss is a Windows-only problem?
The 2.4.20/ext3 issue last year was just a figment of my imagination?
My single hard drive will never fail because it's being controlled by Linux?
This is fantastic news for Linux users everywhere!
--
the strongest word is still the word "free"
The purpose of a raid card is improved reliability. Thus, you shouldn't try to cut corners on quality control to save a few bucks. After all, you're already springing for two drives.
I have had very pleasant experiences with 3ware controllers, both under linux and Win32. Currently I have a 6400 running under Suse 7.3, and three 7400s running under RedHat 8. With some hotswap drive bays, you can even unplug a drive with the system running.
They might cost a little more, but they're widely used under more grueling conditions than the more dirt cheap designs. Also, if you're simply doing raid 1, you can use one of their previous generation cards with no performance penalty, and save a bunch of bucks.
If you're just going to do drive mirroring and you want hardware version and it done well, then you want to go to http://www.arcoide.com
Transparent, no OS drivers required, plugs right into ide, new firmware upgrade to make it so it can rebuild drives w/o software and it works friggin awesome.
It would be useful (to me!) for replies to mention which cards have hardware alarms (audible or visible) and software alarms, and whether these alarms worked and were noticed in the event of drive failures.
I'm considering a mirroring controller for linux, but it's only good if you are notified when there is a problem!
First of all, I've been using some Promise integrated controllers. Imo, Linux support is "okay" but not too good. I could, after some digging, find a driver I could compile for 2.4 series kernels, but documentation was lousy and the driver has been a bit unstable at times. Nothing major, but imo, too much. As it's just an unknown device to lspci, I can't remember the exact chip model, but it's on the new MSI 865PE motherboards.
We've also been using some of these, a bit more expensive Adaptec RAID 2400 controllers. For these, I'd say "stable mainstream" is the right term. Performance is good but not excellent, support is good and compatibility good. No surprises either way and to me, that's good for a hardware device that needs to be trustworthy.
Highpoints, I've only had experience with their older (up to DMA100) controllers and some integrated ones. No good experiences with any one of these, Windows or Linux. Too commonly, when one drive fails, the whole RAID fails. Performance is lousy. I hope it's just one persons bad experiences, but there sure are a LOT of these bad experiences for me.
I have a highpoint RocketRaid 404 Controller and it's working just fine as a controller. I'll hook up more harddrives as funding allowes and use raid 0 But i don't see why that should be a problem. I do think what they mean is that the software they wrote doesn't work with the controller yet. So you have to use it's bios. Boot times to detect 8 drives takes some, but not too long. It runs great AND is fast. Not too CPU dependant, wich is nice on a celeron 466 server. I get about 80 to 90% CPU usage copying 15mb files using mc
Note that all of the controllers listed have one common quality.
They are a garden-variety IDE controller with some extra BIOS and driver code. They are charging you extra because you are willing to pay a premium for it.
Unfortunately, Windows XP and Windows 2000 workstation both ship with the mirroring code disabled, which is unfortunate.
Having said that, there's very little benefit in worying about if the extra BIOS and driver code is going to work under Linux. As long as the chip lets you get to the drives, don't sweat it.
Gentoo Sucks
Open Source, blah blah blah, better quality drivers.
We've been through this discussion many times. "Many eyes, shallow bugs" and all that crap.
But the reality of the situation is that Linux drivers as a result of being mostly Open Source have higher quality than their Windows counterparts.
By higher quality, I of course mean that the drivers put less stress on the disk itself, thus prolonging the life of the HDD. Santa was right on about Linux drivers being solid. I don't know if I'd want to run a 500 client network without backups, but it's true that OSS drivers are generally better than the same device driver in Windows. Maybe it's that all the little secrets of the devices aren't known, so the driver writers don't use the speed-up tricks that Win drivers use (which eventually add excess wear and tear to the device).
When you've got moving parts, the less stress on the device the better.
Damn, it looks like another AC who replies to a troll without a good subject line goes ignored. Apparently, Linux is gona suceed at beeting Microsoft.
aside from read performance, you use RAID 1 for high availability (reduce MTBF, eliminate MTTR from the equation). but already, RAID 5 gives you more h.a.
In addition, it seems like disks are no longer the weakest link in the MTBF chain - assuming redundant PSUs, your main concern is really your CPU/RAM/bus/chipset. To improve this situation, you need n+1 redundancy, and that in most cases means a specialized piece of hardware (h.a. NAS) or a SAN that can be failed over.
If you like that minty-fresh open source, off-the-shelf hw feeling, check out drbd or help out Hr. Reisner. drbd is a linux module that provides for a mirrored block device. If it succeeds as a project, it would be a great way to provide n+1 cheaply.
I use a low-end 3ware raid controller doing raid1 (the 7000-2 product). It has worked great so far under linux for me.s p
http://www.3ware.com/products/parallel_ata.a
I believe it's hardware based, so it will cost a little more than a cheaper software based card(like twice as much-- around $120). I don't have the experience with a drive failing, but setting it up was painless. All you have to do is like press a key during boottime and it's like partitioning a harddrive, but easier.
Check out this forum post here http://www.linuxquestions.org/questions/showthread .php?s=&threadid=53681&highlight=hpt37
And how the west was won. Was a hell of a fight tho.
I use 3ware cards and like them. The drivers are GPL unlike the Promise controllers which, if I recall correctly, are binary only.
Prevent email address forgery. Publish SPF records for y
IDE mirrors are theoretically wonderful, but don't base your entire recovery plan on them. Here's the story from about a month ago... Small webserver running on a Soyo DRAGON plus MB (Promise IDE Mirror built in), 1700XP CPU, 2 60GB in RAID1, Gentoo, etc. I set the system up in January of '02, and it took a digger this past July. Down hard, with a HD failure. When I went to check the contents of the hard drives, I discovered that one was toast (physically defective), and the other had a fully functional image of my system... from January of '03! The mirror had come appart six months ago with no errors, warnings, notifications, or anything else. I'd even rebooted the machine a few times over the last six months (new kernels, etc) and it never glitched or beeped at me. Nothing amis in the log files. No warning.
If u can afford it and need more then 2 drives, definately go with a 3ware card. They are simply amazing. Also they support RAID-5 wich is definately better then anything else, and have an excellent linux support, built-in the kernel. Also they offer cards wich supports up to 12 drives, wich can give you ALOT of storage. I am also currently using a Promise card integrated in the motherboard (a Gigabyte GA7VXP), under slackware 9.0 (kernel 2.4.20). Enabling it in the kernel was easy, but it's not supported by the raid.i kernel. I hooked 2 drives to it and so far so good. I hope that helps
I have an SX4000, and my main problem with it is that they don't keep up on Linux support. They only recently released Redhat 8/9 drivers. For the longest time I've been forced to use Redhat 7.3 in order to use the hardware RAID capabilities. Granted, they did actually release the drivers, but back in January (when I got the card), Redhat 8 had been out for some time. When Redhat comes out with a 2.6 kernel version, I'm willing to bet you'll be waiting a year before you can upgrade. But on the upswing, hey, it does work.
In addition, I really wish they would offer wider support for their chipset in regards to Linux, as you're locked into using specific versions of Redhat as it stands now. I would really like to get Slack on this machine.
I've always wondered why you need drivers for RAID...
AFAIK you build your raid array for say mirroring using the bios of your raid chipset (promise for example). Then it makes sure that any write to hda is actually also a write to hdb. So I think it's purely a bios issue and nothing for the OS to do...
Atleast, that's how I thought it worked, can anyone offer me some insight on this matter?
I am a viral sig. Please help me spread.
Specifically, the RocketRaid 133 (based on the HPT372A chipset). The card runs quite well under Linux. The Linux driver from HighPoint is quite good (sadly, only partially open source) and provides a /proc interface. (Don't worry, it compiles just fine in any kernel.) Sadly, I cannot show you output from the interface because Slashdot refuses to let me post it (citing junk characters). Stupid Taco. The interface also allows you to issue commands to the controller without rebooting, but documentation is poor. The BIOS utility is also quite good.
Anyway, the device hosts two RAID-1 arrays, one with 2 80Gb Seagates (ST380021A) and the other with 2 200Gb Maxtors (6Y200P0). They appear as SCSI devices. I have tested the mirroring and I am mostly satisfied.
Basically, I simulated a failure on one disk (removing it then performing some work on the other). When I reattached the drive, the card recognized the "failure" immediately and wanted to build the array. For my first test, I let the BIOS do exactly that. Took a very long time, but the mirror was recreated successfully and there were no problems (I tested by removing the first disk and trying again--the mirror was good). The second test was letting the driver do it after the machine had booted. This was a dismal failure. The card does NOT like rebuilding the mirror once the system is running.
Performance is quite good. Even though this is not HighPoint's latest offering, I am still quite impressed. I don't have any hard benchmarks, but I can post some later if you'd like.
These two arrays are accessed by many machines in my home network over NFS and by on average 5 users logged in remotely. They serve games, web pages, and my software, movie, and music archive. These arrays take a decent amount of stress, but nothing severe.
I'll post more in this thread if I think of anything. I'll answer any questions about the card's performance if you have any.
Join Tor today!
With RAID-5, you can lose one drive and still retain your data. However, there's always that chance that you'll lose a second drive before your spare gets completely rebuilt. In that case, there's no way to recover your data. (Correct me if I'm wrong.)
Let's say, then, that I want to make a RAID-1 mirror of two RAID-5's. Two racks of drives, each rack in RAID-5 configuration, and I want to mirror those racks, so each set can lose a drive, and/or one rack can lose many drives, and I'll still have my data. Ideally I'd like two controllers, so if a controller fails, I'll still be able to access the data from one side of the mirror.
Is this possible, and if so, does the mirroring of those two controllers have to be done in software? Or is there a RAID controller controller that would allow that kind of operation?
I think you're missing the point, and blatantly trolling...
Holy crap, what an zealot. Could you even read your own post without laughing?
Really? My disks spin constantly unless I issue an hdparm call to tell them to sleep. Thats the default behaviour out of the box, at least for slackware 9. Though frankly it doesnt matter if they're spinning or idle.
I've seen harddrives fail in linux, windows, os/2, macs, xboxes, creative nomads. If it moves it can break. Windows doesnt add any more wear and tear.
If your harddrive is going to fail, the parent is correct, its a mechanical problem that doesnt care what OS you run. Linux does not make your disk last longer, and I defy anyone to prove otherwise.
This is a huge topic, and I've only ever looked into part of it in depth, but:
Remember that for most motherboards, their RAID implementations are going to be software based, and not hardware. This makes for a cheaper chipset in general, and since the CPU speed is fairly high now the performance hit you experience is pretty nominal compared to the performance of a hardware based controller. The only REAL exception to this is for RAID 5 controllers - RAID 5 uses an XOR system which is really punishing on a CPU - hardware (i.e. add-in cards) makes a huge difference here.
All things considered, bear in mind that one of the reasons for RAID is data integrity (depending on mode, obviously) - do you want to risk your data by saving $50 on a cheap board? Where I live, the hardware place I frequent has a fairly large (10+ on staff at any one time) team of techs assembling custom orders and doing installs of purchases if their customers want them. These guys are usually the ones fixing/replacing problem MB's, and they'd be the best to talk to on quality - most reviews of MB's nowadays tend to focus on features and performance - long term reliability is something you've got to find out from others or on your own...
-- "We are all in the gutter, but some of us are looking at the stars" [Oscar Wilde]
I'm using a 3ware 7xxx with 8 160G Maxtor drives and I really like it. I'm using it in RAID-5 mode, but it supports mirroring as well. The only problem is the Maxtor drives are a bit on the slow side. When it's time to do it again I'll use faster drives. With the large drives available now, it is easy to hit the 2^32 block limit.
http://www.cardaysupply.com/products/displayProduc t.asp?id=19
Lots of promises, are they for real?
I've experimented with IDE-RAID cards from Promise, SIIG, and Duplidisk.
I particularly like the Duplidisk (even though it's the most expensive, e.g. $200?) because it works at the IDE 'level' and is totally transparent to the OS. The only limitation is that there's no DMA support unless you have two identical HDs; I suspect the other cards work around this by including their own device-drivers (but this is OS dependent). Duplidisk also has great tech support.
Another issue for all these cards is that they can't/won't work around bad sectors. I suppose this is a Good Thing but it was disappointing to me as I have about a dozen used HDs at my office that have bad sectors but otherwise work great. This is also a possible source of confusion as the cards will report "Dead HD!" if a bad sector appears on the drive -- the drive doesn't have to actually die for it to be rejected.
Not offtopic (I hope), given the original poster's minor slam on ECS.
/. Wow.
I've been nothing but plesantly surprised by the ECS K7S5A/Athlon XP combo, because:
1. SiS used to suck as a chipset manufacturer, and
2. ECS used to have REALLY bad quality control.
The surprise? SiS has one of the most stable chipsets anywhere (not the best performing, but performance is irrelevant if the drivers are wonky). Add to that the fact that I've worked with about 10 boards (all bought at different times), and I've yet to have a single problem with them under Linux or Windows (in fact, the SiS chipset seems to perform better under kernel 2.4.x with the SiS chipset than even Via). I've standardized on that combo at home and with the boards I've replaced at work.
Summary: ECS has come a long way, and SiS even farther in the last couple of years. What the boards lack in performance, they more than make up for in stability.
Geez. I never thought I'd be an SiS advocate on
"Time flies like an arrow; fruit flies like a banana." --Groucho Marx
I think these cheap RAID solutions don't deliver great performance gains, and usually they deliver great driver headaches. Under Windows and Linux.
As I have tested, a good pair of disks with software RAID (specially under Linux) can deliver near the same performance. Without the hassle to match firmware versions with driver versions. No more unstable systems. No more corrupted data. I had very bad results with Promise and Highpoint controllers. Specially with RAID 0 setups.
If I need performance or security, I'm using true SCSI RAID controllers. Sure it's a high investment, but they're worth every $$. Maybe it's the same with high-end ATA RAID controllers, but sure not with the low-price cards or embedded MB controllers.
What a bunch of non-sense. RAID won't prevent any wrong data to be written to the disks. It justs makes it against hardware crashes. If the controller or other sub-system goes awry, you can still lose your data. Operating systems do not have any influence on the controler (or does any other software, like virusses). If a malicious piece of software destroys your data, it is destroyed no matter what.
The thing I hate about those controller cards (both ATA and SCSI is that they use a significant amount of time to boot up. Especially if you've got RAID incorporated on your motherboard, this is a serious drawback.
With the current state of hard disk drives, RAID 1 makes sense though. Failure is always imminent it seems. It seems about time that hard drives are replaced by other means of storage without moving parts.
Unfortunately, those other memory technologies do not seem to catch on significantly.
Warper
ps. currently suffering from a SMART warning on my WD 120 GB drive. Darn. Let's check if I've still got 2 years warranty on the thing. I think that that is the minimum European warranty anyways.
This is a topic that hits home for me. I have experience with Highpoint, Promise, and 3ware controllers. I used to use a Highpoint chip embedded on my motherboard. It performed fairly well, with the exception that it took two IRQs, and didn't allow my soundcard to function properly. I would get all manner of pops and clicks in the sound. It eventually drove me to the point that I bought a new motherboard and a Promise card. Ah, the Promise TX2 1000, how I hate thee. I will never buy a Promise card again. I spent more time rebuilding my system due to data corruption than I did using it to get work done. (Exaggeration) It caused unrecoverable drive corruption on three occasions in less than nine months. I switched again. Now I've got a 3ware 7506-4LP. I love it. It's a 64-bit 66MHz card, does RAID-5 and RAID-10, (in addition to RAID-0 and RAID-1) and it's fast. I've had it for a little over a month now without any trouble. Yes, it's more expensive than the cards listed, but I think it's money well spent.
Is it a troll, if he consistently argues for the same position?
I'd say he was an FSF zealot. Those are frequently mistaken for trolls because of their complete disconnect from reality.
You actually use Promise cards for important data?
Spend a few more dollars and get a 3Ware card.
They are fast, reliable, have low CPU usage and are supported under Windows, Linux and BSD.
Please put this little article on your resume so that anyone with a clue can avoid hiring your cheap ass.
If tyranny and oppression come to this land, it will be in the guise of fighting a foreign enemy. - James Madison
I haven't played with RAID's other than RAID5 much, but here's my experiences. Maybe they'll help your opinion.
:)
/proc/mdstat .
I've used several different external RAID5 solutions, including the Promise TX8000 and Radion arrays (and a bunch of no-name brands too). Those all worked beautifully, and survived failures without any significant problems.
I've worked with four internal hardware raid solutions. Two were the Adaptec AA-130 cards years ago, and the most recent two were Promise SuperTrak SX6000. One belonged to a friend, and one went in my machine.
My friend had several problems with his. I'll leave his story at that.
I put together a nice dual processor machine for a large photo archive site. It needed lots of space, which is why we needed the RAID5. It also needed to take up as little space as possible, so we opted for putting 6 drives in a 2u case with the SX6000. The card was incompatable with the newer chipset of the motherboard. It took two weeks of daily calls to tech support before we gave up. A week later, they released a firmware update which addressed this problem. The SX6000 doesn't handle heavy read or read/write traffic very well. When we made it an active web site, the server would crash very frequently with errors about the array. If we kept the traffic slow, it worked fine, but that what we wanted from this machine. This array solution proved to be non-functional for us, so we made the machine a backup machine, so now we have like 600Gb of storage space to back servers up to.
I've heard a lot of good things about 3ware, but haven't tried one myself yet.
I've been experementing with Linux's software RAID's. I've used RAID1 and RAID5, and they both work great. I've had a drive fail on two so far, but for the number of drives we use, that's acceptable. They rebuilt fine sticking in a new drive, with very little performance hit. I do like that Linux gives decent statistics in
You seem to want multiple platforms, so I guess Linux software RAID isn't much of a solution for you.
I can simply warn against the Promise SX6000. It should have been a good card, with Promise's reputation, but I was severly disappointed.
Serious? Seriousness is well above my pay grade.
I've been using this motherboard for quite a while and I like it, the HPT 370 chipset works fine as plain IDE ATA100 driver, I'm pondering buying some "twin drives" though and setting up some RAID. Anyone with some experience on setting up RAID on that hardware?
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
I don't know how much insight I can provide, as far as summarizing the "big picture" of RAID controllers in Windows. I haven't ever purchased any of the newer Adaptec RAID cards for use with Windows 2000 or XP, and the only ACard product I've used recently was an Ultra-66 EIDE card for the Mac.
I do, however, own an MSI (Microstar) Pentium 4 motherboard with on-board Promise EIDE RAID support, and have been using it for quite a while now.
So far, I guess the product basically works "as advertised", in that it's doing hardware RAID and hasn't screwed anything up during disk writes.
Performance-wise though, yes, the boot time is very lengthy. Not really any more so than the older Adaptec SCSI cards were, however. If you're used to having Adaptec SCSI on your system, you'll find this feels pretty familiar from the boot-up standpoint.
I also have a concern that may or may not be founded in reality. (Comments from other Promise EIDE RAID users are welcome here?) When I first set up this board, I used a pair of brand new Western Digital 100GB 8MB cache drives. After only a week or two, I started having problems where the disk access light would stay lit during the middle of a disk read, and the system would freeze up. Occasionally, this would result in a blue screen of death. From listening closely to the sounds the drives were making when this happened, I decided it was a disk error (bad sectors), and swapped out the drive that appeared to be the culprit. Not too much later, this started happening again with the other drive. At this point, I swapped both out with the 120GB version of the same hardware. Out of curiousity, I reformatted both of the 100GB drives and tried using them as regular boot drives in other systems. Both worked without any problems!
Meanwhile, it's been at least 6 months since then and the pair of 120GB drives are still doing RAID in my P4 without any real issues, but every once in a while (maybe once a week?), I notice it sort of "hang" for a couple seconds doing a disk write, and hear that familiar drive seeking noise like the drive has a bad sector. Then it seems to recover and everything is fine again.
All of this is making me question whether or not the Promise on-board EIDE RAID has some bugs still in the firmware, or slight timing issues that can cause drives to appear to be malfunctioning when they're really not?
At the very least, I would have expected some sort of report from the Promise card that a drive was failing, when I had those problems with the 100GB drives.... not just a blue screen of death from Windows XP.
Added Information:
I wrote the Slashdot story. I've gotten additional information since then:
First, I don't see any evidence of low quality in ECS motherboards. Both ECS and Fry's employees say that the high rate of returns is due to the fact that the low cost attracts people who are very inexperienced at building computers. That seems right to me. An ECS sales representative mentions that Fry's uses ECS motherboards in the computers it builds. The ECS motherboards are suitable for business use. They may not be the best for overclockers or gamers.
I wrote a long letter to Promise Technology Technical Support about the fact that Promise is allowing ECS motherboards with Promise RAID mirroring controlers to be sold without software that is necessary to monitor the health of the mirror. The letter is below, and represents my opinions at the time. Even though the national sales manager of Promise suggested I send the letter, there was no answer.
Below the letter is all information and opinions that I had available at the time.
___________________
TO:
Kevin Hong
Promise Technology Technical Support Engineer
Kevin,
Jason Turk told me to write to you about a problem with Elitegroup's L7VTA Rev. 1.0 motherboard, which includes a Promise RAID controller chip.
Elitegroup says they have never been provided the FastCheck or PAM monitoring software for this motherboard. Can you supply monitoring software that is certified by Promise to work? The software is not on the CD supplied with the motherboard.
We need not just software that appears to work, we need software that is certified by Promise Technology to work with this motherboard. Several people at Promise have told me that the software provided on the Promise web site for the Promise RAID controller adapter cards may not work with the OEM controller chips installed on motherboards. I suspect this is not true, but merely a Promise Technology marketing scheme. However, we can't risk selling these to our customers and finding later that there is some hidden serious defect.
Thanks,
Michael Jennings
Futurepower Computer Systems
_______________________________
Complete Information:
So that all the information to resolve this issue is supplied in one place, I have provided everything that I know and think below:
Some versions of the Elitegroup L7VTA Rev. 1.0 motherboards are being supplied with Promise Technology's PDC20265R IDE RAID controller. Elitegroup supplies three of the four elements needed for a RAID controller. The motherboard and included CD have 1) Promise Technology's PDC20265R ASIC chip, 2) the latest version of the BIOS, 3) and the driver software.
However, item 4 is missing. Elitegroup says Promise Technology never supplied RAID array monitoring software for this chip. Promise Technology calls this software the "FastCheck" or the "PAM, Promise Array Management" utility. Without this software, the user cannot know if a hard drive has failed in the RAID 0 or RAID 1 or RAID 0,1 array.
This issue is Elitegroup's Case number RAE54616. (Note that Elitegroup also calls themselves "ECS".)
Elitegroup's web page for the L7VTA product is here:
http://www.ecs.com.tw/products/pd_spec.asp?product _id=327
The L7VTA V1.0 driver page is here, showing the latest Promise drivers:
http://www.ecs.com.tw/download/dw_spec.asp?product _id=63
Promise Technology's web page for on-motherboard RAID controllers is here:
http://www.promise.com/product/oem_ataraid_pdc2026 5r_eng.htm
Note that this Promise web page mentions all four elements of a
Everyone knows Linux is quite a bit more stable than Windows. I really don't have any need to do this kind of redundant replication.
/" as root on your Linux box, again,
kiss your data goodbye regardless of
mirroring.
Assuming you didn't mean that as a troll...
HDD mirroring generally provides protection from *hardware* failure, not software glitches. If Windows suddenly decides to overwrite your porn collection, a RAID controller will faithfully do it on both drives. If you accidentally do an "rm -rf
Specifically, you gotta watch for the following type of weird case:
This is not a common scenario (it happened only a few times with hundreds of systems over a few years), but it's one that's pretty hard to test for. You can have controller firmware bugs that stay hidden for years before they popup. Sometimes the firmware works fine, but the sys admin gets confused with all the garbage errors flying at him/her and chooses the wrong disk as the primary.
My experience with this was on relatively large HP/Sun/SGI database servers.
Hardware/firmware mirroring is great the vast majority of the time. But if you're a truly paranoid sys admin, I'd recommend loosely coupled hot spare replication.
When I say loosely coupled, I mean don't replicate low level disk block by disk block, since that can just replicate possible corruption. Instead, replicate at the logical filesystem write level.
If you're really really paranoid, the replication queue should have a few hour delay so you can cut if off if an application level error deletes massive data (but where you can quickly flush the queue in a failover situation)
I'm not a sys admin, but a database weannie, so I don't know which Unix level products are currently good for this.
Ben in DC
"It's the mark of an educated mind to be moved by statistics" Oscar Wilde
joysticks.
Hmm a mirroring controller...
A mime controller? A monkey-see, monkey-do device?
Or some kinky sex gadget. hmmmm.
I bought a Promise tx2000 for home because they claimed to have linux support. It turns out that (a) it's binary only; and (b) it's only for 'certain' kernels, and much later than the distribution updates. Install a new kernel, lose your disks! I ditched it and used the software raid in the kernel. Now everything is much more solid and I feel better with my data. I definately wouldn't buy promise again. It's a windows company with limited interest in linux.
they make good stuff.
I have a 7000-2 (about $100) card in my bsd box. linux support is even official I think.
As you can see, these controllers cost from $59 to $85. Last weekend Fry's was selling a retail boxed 1800 MHz Athlon XP processor with fan included and an ECS motherboard for $59 total. So $80 seems like a lot for a little card with one chip and a flash ROM.
Damn, we paid about $500 for an UltraSCSI 320 RAID controller and you think 80 bucks is expensive? And each one of the 75 GB drives cost $550, ouch!
I guess you allready figured out that I work for the govn't. 8) It's not the same planet as the private sector.
...adapters in seperate computers, obviously.
The promise is on the MB of my p4t533, and it only supports two drives, either R0 or R1; I run mine in R0, for the additional speed. I make backups of critical files on other machines for safety. No failures, and damn fast with two 80GB Maxtor 7200rpm drives.
The rocketraid and IWILL are both highpoint controllers, and the only difference I can tell is the Iwill is ATA-100, and the rocketraid is ata133.
There is no real difference in the performance of the three cards; the drives only transfer about 33Mb per second continuously; regardless of ata110,133, whatever. Big files empty the cache, which is all that ever transfers at the higher speed. (Many sessions with various benchmarks support this.)
Now for a real raid card, I've got an IBM monster that can handle up to 45 hot-swap drives; I'm ordering drives (5) to set up a raid 5 array soon, and I think that the scsi U160 will totally kick the IDE stuffs ass. But we'll see.
The cool thing about using the raid array is that UT2003 starts in a few seconds, and everything else just pops up immediately.
Truth isn't Truth - Guliani
Sex Sexy Sexiest.
If you have a serious RAID-5 configuration, you should have alarm bells going off and sysadmins getting paged in the middle of the night if a drive fails.
So:
a) See how long your drives have been running since they were first installed.
b) Look up the mean time before failure for the type of drives you are using.
c) Estimate the amount of time it will take your sysadmin to get to the hardware and replace a drive if one fails.
Using those numbers, figure out the odds that you will have a second drive fail before the first dead drive can be replaced. Most of the time, we're talking Insanely Astronomical Odds.
And if you are running a mission critical system, you probably back up the data to another media format on a regular basis anyhow. So you might lose at most 1 day's worth of data, IF lightning struck in the same place twice. Or whatever.
Short answer: you really don't need dual RAID-5 setups.
On the other end of the spectrum, however, is OS-level mirroring. I have only experience with Linux RAID-1, both that has never failed me either. I have run software RAID-1 for many years and have only (long ago) had problems where I had to "raidhotadd" a drive from a mirror set that was no longer recognized.
My recommendation, based on both performance and ease of use, is to use a 3ware controller. That coupled with Linux LVM makes disk management and resiliency a breeze.
My only disappointment with the 3ware controller is that one cannot make take a single disk with data on it, add a second disk and tell the controller to mirror the first disk to the second. With the 3ware, you will lose the data on both disks when making a mirror. (So says 3ware tech support.)
the growth in cynicism and rebellion has not been without cause
I tried building a RAID 0+1 system with a RocketRAID 133 on a dual athlon system running linux. I ran into nothing but problems:
- The card crashed at boot. Display corruption. Luckially a BIOS flash fixed that.
- The card doesn't support booting Linux. At all.
- The card's drivers for linux are horrible. It took me 2 days of fiddling with compile settings and kernel settings, and all sorts of settings before linux would even recognize the card.
- Once I got the card running, Sybase would crash out hard, taking the system with it, when trying to load in a largish database (3gb). I'm not sure if it was because of the HighPoint card or because of the Athlons, but it just didn't work out.
In the end, I gave up, sent all the parts back to newegg.com and bought a Dell with proper SCSI RAID.
Adaptec's 1200A is Highpoint card with a slightly different bios on it. Heck I've heard you can flash it with highpoints firmware.
/ 79 7261
http://www.internetnews.com/storage/article.php
IANALBIPOOGL (I am not a Lawyer, but I play one on GrokLaw.)
I've had exactly the experience you describe. When there is a hint of trouble, it is necessary to break the mirror and test each hard drive separately.
I use a promise Raid controller on the motherboard. As promise suggests, I use the synchronization feature from time to time, although as the author suggests, it doesn't really make sense that you should have to use this. I'm wondering what other people's experiences are with synchronization, and whether or not this is actually necessary, or even a good idea.
I should note that the way that synchronization appears to work is by reading data from one drive, and creating a fresh, bit by bit copy onto the other drive. This seems somewhat dangerous, since if the source drive happens to have corrupted data on it, then the data on the destination drive will also become corrupted, when it was not before.
Instead, consider going with fully managed SAN storage. Yes, it's immensely expensive, but then again you never need to worry about hardware failure-- the SAN will contact a technician and tell them to come fix it before it breaks. (EMC makes a killing on this sort of business)
Again, you're paying a heavy premium for this type of service, but if you feel that RAID level 5 isn't reliable enough for you, you're going to wind up paying through the nose for absurdly complex hardware solutions-- you might as well pay someone else to worry about it for you.
There are binary drivers for Linux, and partial source drivers that do raid. Of course they are a pain much like Promise's binary only drivers. Highpoint actually recommends you run the card in JBOD (Just a Bunch Of Disks)mode under Linux, and use the md (software raid) driver. Keep in mind all of these cards do raid in the driver not in hardware.
IANALBIPOOGL (I am not a Lawyer, but I play one on GrokLaw.)
alias rm=rm -i
It can work, but be extremely carefull, I never could keep Grub from breaking the mirror, and you half to tell the boot loader not to probe the ide drives - because the controller does SCSI and ide, but if you probe IDE it stops mirroring right and partition labels get confused. I did get lilo to work, but none of the utilities. You also need their closed-source hpt linux controller, which IMHO is picky and a pain to install with.
Promise's binary driver sucks. Why you ask:
1)You are stuck to a handful of kernel, and Promise take forever to release drivers for new releases of Red Hat. God help if you don't run Red Hat.
2)When you lose a drive you system crashes!! (Yes I've tested this.)
You are better off using the controller as a cheap ide controller, and running software raid under linux. Just be sure to use lilo instead of grub as grub doesn't understand software raid very well.
IANALBIPOOGL (I am not a Lawyer, but I play one on GrokLaw.)
nazis are cool. bush-followers are just pale copies and wannabes.
All of the cards listed are a joke. They are "semi-software" RAID cards. They are crap compared to the real deal, such as cards made by 3ware and the like. Cards with actual processors/cache/etc. on them, and not just some hook into a hard drive interrupt.
If you want to go cheap, you're much better off using Windows 2000/XP's dynamic drives/mirroring ability in software.
I never will understand why motherboard manufacturers use Promise/Highpoint chipsets on their "high-end" or "server" boards. Complete waste of time.
Not All Who Wander Are Lost
I'd say my experience is limited to the 'fasttrak' raid controllers from memory. I wasted a good two days getting this working only to find them unreliable on Linux. They were useless and wasted about $700 AUD (including their proprietary 'superswap' drive bays.).
My advice, until they really start being open with Linux. Don't go there - it's money down the drain.
alias rm=rm -i
...Evaluates, under the user issuing
"rm -rf /" to "rm -i -rf /".
Since the latter of a pair of mutually exclusive options takes precedence, the "-i" has no effect.
can someone explain what each RAID specification means exactly? Although I know RAID 0 through 5 exists I'd like to be clear on the distinctions of each.
Software RAID (under Linux) is a better deal than spending $60 on a RAID card. The RAID drivers for these cards basically just implement software RAID anyway, so what's the point of having the hardware? The card is just one more thing to go wrong. Use the IDE controller that comes with your mobo.
Software RAID has many advantages. You can monitor everything about the RAID array from within Linux. Drivers are no problem. It is well-tested and stable.
Also, you can get better performance. I would rather spend an extra $200 for a dual-CPU mobo, get a second CPU, and run software RAID, than spend $200 to get a dedicated RAID card.
Unless I'm mistaken these cards don't do RAID anything. Indeed, they are ordinary controllers that have a few extra functions, all of which are basically hardware acceleration for RAID operations. The critical bits of the RAID are still done in software.
With that said, if you're going to do software RAID, you might as well just do it. I've been using VINUM on FreeBSD for a long time, there are other options for FreeBSD and for Linux. The better ones take full advantage of the "dual read" property of mirrored drives at the kernel level. Best part, no special hardware required. If something breaks buy any old ATA card and you're back in business since it's all in software.
If you need RAID-4 or RAID-5, a dedicated card might be a good idea. For mirroring I think software is superior.
3Ware controllers are expensive, I note. The cheapest: About $130.
You said, "My only disappointment with the 3ware controller is that one cannot make take a single disk with data on it, add a second disk and tell the controller to mirror the first disk to the second. With the 3ware, you will lose the data on both disks when making a mirror. (So says 3ware tech support.)"
That's a BIG disappointment. The on-board ECS Promise controllers, as well as the adapter cards have that.
Anyone else think the reference to the 80s arcade game is cute?
My company has had good results with DupliDisk controllers. The nice thing about them is they are totally transparent - the OS doesn't even know mirroring is in use!
...I've been running it for 6 months under Mandrake 9.1. Smooth sailing...I've had RAID 0 and 1 arrays on it without any problems. Also supports 1+0 and JBOD. The RR454 is out now, too, which supports RAID 5 (ooh, if only I'd waited...) Great value, though--can run 8 drives for ~90 bucks. Hit up pricewatch or something, they're easy to find.
Really? My disks spin constantly unless I issue an hdparm call to tell them to sleep. Thats the default behaviour out of the box, at least for slackware 9. Though frankly it doesnt matter if
:) )and just doing simple maneuvering makes the disk access heavily. I run Linux, and the memory is used for the buffering, and data is written in a very optimized fashion.
:P
they're spinning or idle.
Disk spin is different than access. Of coure your disks spin until you put them to sleep.. otherwise you'd have access times in the seconds.
The truth of the matter is that Windows does access (read: access, not SPIN) the disks more. On this same machine I run Windows 2000 dual boot. (Command & Conquor Generals, need I say more?
No zealotry, just plain hard fact.
Sure hardware fails under all OS's, it's just that most OS's don't try to beat it down like Windows
-- This space for lease, low setup fee, inquire within!
It is possible to get it to run (via a heavily hacked) 2.4.20.
In other words this is *NOT* reccommended (and I have gotten no response from the maintainer of the package (aec62xx) )
I have had a some experience with Promise cards, namely the Fastrack100 raid series, on the Win32 platform.
:(
:)
Generally they have performed reasonably well, with little hassle. However, there have been some major exceptions/issues:
* BIOS and driver upgrade -- BIOS upgraded A-OK, but the driver achitecture changed slightly between revisions (ie. one dll disappeared) and Win2K just refused to load the new driver (kept asking for a dll that no longer existed), or revert to the old driver. Joy. Spent the weekend rebuilding the server. Promise technical support was useless.
* I am yet to have Norton Ghost work properly with my RAID mirrors. The copy gets to about 99% and dies. Googling hasn't helped much either. Thus to do a ghost I have to disconnect one of the drives from the Promise and stick it on the regular IDE ports. It generally helps to remove the Promise card entirely so it doesn't throw a wobbly about the mirror being broken every time the box is reset. Joy. Mucking around in cases is fun
* The manual is generally poorly written. Explanations are unclear, and important information can be burried.
Other than that, go for it
Previously on slashdot there was a story about an article from The Tech Report that is really helpful regarding making a decision about IDE RAID card price/performance. From examining the article, it is obvious that 3Ware comes in with a better card for the money.
Okay, I know I can't be the only person this happened to. Promise supports only the stock Redhat 7.3 kernel, not the errata kernels. The stock 7.3 SMP kernel shipped with a bug that made the system friggin crash (remember that?). So what do you have then? A server that crashes and can't be fixed because of the stupid proprietary RAID card driver. The lesser of evils was to not use the SMP kernel and just run off of one processor. I thought that was going to be a temporary fix, but it has been over a year now and Promise hasn't released updated drivers. All they would have to do is open source their stupid drivers and they would sell sooooo many more of those little cards. Damn.
My company builds ~8K systems a year. I recommend 3Ware (first choice) or Adaptec for ATA RAID. You pay more, but you get what you pay for. I've sold dozens of systems with 3Ware.
This brings me to my 2nd point. Who else stopped reading after this part below? Excuse me if I ignore your quality assurance recommendations based on 3 systems:
Employees at Fry's tell me that ECS has the poorest quality control of any of the motherboard manufacturers that they sell. To me, ECS motherboards seem to have surprisingly high quality. However, we have only tested three, and only one extensively.
Stay as far away from Promise as you can. Their support is by far the worst I have seen.
We paid about $10000 for a Promise RAID solution about 2.5 years ago (UltraTrak100 TX8). Lately we had problems with it. They said, "We only support up to Redhat 6.2. Sorry!"
I expect better customer service after paying that much.
On the other hand, we have used 3ware for several different IDE RAIDs. Not only are they the only ones to have kernel support for **HARDWARE** IDE RAID, their support so far has been top notch.
I've been using a FastTrak-100 controller under W2k for about 3 years. I run two mirrors w/ two drives each, one the OS boot disk and one a data only disk. In general I have been very happy with it. I've had one drive fail and doing a cold swap was painless.
When I first installed W2k, some standard software (Roxio Easy CD Creator was one of them), and then the FastTrak driver I kept getting bluescreens pointing to the FastTrak driver. By chance I figured out that it was some sort of conflict with a backup program in Easy CD which I removed and that solved the problems (This backup program is no longer included with Easy CD).
I like the Promise, their website seems to have gotten worse with time, but otherwise the card has done the job and continues to do it. I will consider other cards when I build my next machine though.
I believe the point was that any corruption which is software based, will corrupt both sides of the mirror pair, or your raid-5, or whatever kind of volume you may have.
So... just because you have mirrored hard disks does not relieve you from doing backups. But if a hard disk bites it, you may not need those backups... this time.
They say the first thing to go is your penis. Well, it's either that or your brain. I forget which...
I would never consider a controller that doesn't do that. It's a critical feature for any RAID card in my mind.
Recently got a FastTrack SX4000, here's my expirence... Wouldn't boot properely on my not that old computer (FIC AZ11E w/Athlon 800). Their support staff bent over backwards to try and resolve the issue. Even sent out replacement card. I had to buy a new computer to get it working, but I have nothing but good things to say about their support.
I've configured 2 servers with Promise cards. They have a "partial open source" module which you can compile to fit your kernel. Worked fine for everyting between 2.4.18-2.4.22
So far, I've found that using the open-source module works just fine.... my only beef being that you have to compile SCSI support (as a module) into the kernel in order for it to load properly. Ah well.
I've used the Promise TX2000 card on a file server with two 120GB drives in a RAID 1 configuration and found an odd problem. After putting the file server on a KVM switch and booting it up, the Promise BIOS inidcated a problem with the mirror. Win2k Server would not boot at this point. It was impossible to tell which drive was failing by looking through the BIOS menus, so I pulled first one drive, then the other to find the culprit. Since I replaced the bad drive and rebuilding the array (took about 2 hours to mirror it again), everything has been running as before. But I was curious as to how bad off the 'bad' drive was, so I put it in a desktop. The machine booted fine with it in, and all the files were visible. I formatted it, ran a chkdsk on it, even used the WD tools to check it out. Nothing wrong could be found with the drive. Perhaps the controller messed something up with the its mirror checks and couldn't recover? I don't know anything about how RAID works, but this seems really odd to me.
"Anyone that has ever gotten an idea based on any of my work and done something better with it-good for you."--J.Carmack
Where does one find the simplest of information about RAID? I design electronics and write software, and this should be simple, but I can find no general use description of what RAID does with the drive. For example, you suggest breaking the mirror and testing each drive. I did not realize the drives were usable after being put into a RAID. Nor did I realize that one could put them back together again. I am learning alot here, and I know this information is almost trivial, but I still can't find a good explanation. I know what each raid level means, but only at some top level. Any info appreciated!
Jeff Long
President
Granite Precision Inc.
jeff@graniteprecision.com
We've been using the DupliDisk2 (and now DupliDisk3) by ARCO... does transparent mirroring between 2 drives. One end plugs into the MB's IDE controller, and the other two ends go to two separate drives. The whole thing is transparent to the system, unless I choose to install their Linux drivers (needed to transparently rebuild in a production machine). They're more expensive, in the $200 - $250 range. But the nicest thing is that you can always take a non-mirrored system, slap a DupliDisk in there (w/ another drive), "rebuild" the new drive and be done w/ it. Performance is good, and they've saved my boat more that once. Anybody else play with these things? Oh yeah, they're at www.arcoide.com, but I generally their stuff from catalog resellers-- better prices.
This was a Promise SX6000 btw and I was doing raid 5 under linux. I'll make this short. I had a drive failure, replaced the drive, the controller would not rebuild the array. Backed up all the data, rebuilt the array from scratch, reloaded everything. Time goes on. Another drive starts making an unusual noise. I happened to have a back up drive handy. I gracefully powered down the system and swapped out the noisy drive with the new drive. The controller would not rebuild the array.
I said screw it and bought a 3ware 7500. Have had it for about a year now, no problems.
I actually just sold my SX6000 about 2 weeks ago for 2/3 what I paid for it. I'm glad I hung on to the piece of crap instead of pitching it.
Stupider like a fox! - H.S.
i've seen a LOT of kt7's where the caps on the motherboard begin to bulge at the top and eventually leak or explode, killing the board in the process. You might want to double check them. Its a well known problem, check google for references and the story behind it.
Lawyers, MBA's, RIAA? A jedi fears not these things!
I am about to buy an Adaptec 1200A because I have hunted around and found people near me who run Linux servers on them. The 2400 has support listed on the web. So I'm willing to assume the 1200A will work for me.
"3ware has on-board logic to handle most of the RAID protocols independently of the main CPU, including RAID 5 and variants. The low-end guys do all this work in their driver, which is one of the reasons the are unlikely IMO to open source them."
That's because they have an on-board cpu, coupled with some FPGA's, and memory.
Intel's ICH5R south hub seems to provide good RAID support and is lots cheaper than the PCI cards./ iaa_raid /
w thread.php?s= &threadid=8156&highlight=raptor
http://support.intel.com/support/chipsets
Sadly no Linux drivers yet, but the performance seems better than other cards due to no PCI-bus limitation:
http://forums.overclockers.co.nz/sho
If you want RAID under Linux, just use the motherboard ATA interfaces and software RAID; you'll get the same results with no driver hassles. IIRC Win2K (and maybe XP Pro) support software mirroring as well.
Happened to me last week. Backups gooood.
Well I have an adatptec 1200A and I found that while it can mirror or stripe disks, it is painfully slow.
I ran some tests comparing disk I/O on my systems built in IDE controller vs the performance of two drives in a striping config on the 1200A.
The RAID config was slower by about 50%!
I contacted Adaptec support with the info for my system config and they responded that they were not surprised at the results. The main point being that the Adaptec card being a PCI card is substantially handicapped vs some IDE controllers integrated on the motherboard.
To their credit, Adaptec has offered to refund my money. But my advice for people is to just use software RAID and avoid these low-end RAID cards, its cheaper and quite possibly faster.
I've found Fry's to have the poorest quality employees of any motherboard retailer!
-- Jack
Ive been using this card 24-7 in an dual Athlon system since December and its been great. I chose to do RAID 10 with 4 drives. Rebuilds do kinda suck, but the 3ware desktop manager allows you to adjust the parameters to make general use acceptable. Probably wouldnt want to spool any video while its doing this, but everything else is ok. I believe I paid about $300 then, newegg has for $275. It also does raid 5. Keep in mind they have a newer version out said to be faster. In anyevent, anybody considering raid should read this
My experience with 3 Adaptec 2400As (RAID5) is not good: #1- up in smoke (literally) after 8mos moderate duty. #2- replaced #1, still running (crossing fingers). #3- DOA (verified by Adaptec). All 3 cards went into Supermicro dual Xeon boards in SM server cases, protected by high end UPSs. I've leared my lesson, 3Ware from now on.....
More to the point, the mirroring only protects against data loss due to hardware failures. I can't count the number of people I've dealt with that think that they don't have to worry about backups anymore because "Hey, we have a mirror, it's protected." {sigh}
The higher the technology, the sharper that two-edged sword.
I bought a 3Ware 6400 back when they were new. Before the 6410 came out, and you had to get a full length PCI card for 4 ports. I got it through hypermicro.com, who I knew and trusted for all my storage purposes.
Until the 3ware, I had been "The SCSI guy" in most circles. I was running all scsi systems starting in 1992 and woe betide he who suggest I add an inferior IDE device to my system. I had first spotted the 3Ware at PC Expo back when they were prototyping the 6000 series, and immediately scoffed at it as "just another IDE RAID wannabe." The 3Ware engineers overheard this and invited me to come over and play with their system and tune the benchmarks on it (standard ones like iometer, not custom in house things you can fake easy) to prove they were serious. I left their booth saying "I want one!" and repeating it to my friends for the rest of the show.
But I digress. I had purchased a 3ware, to go in my new Athlon system. This was back in the old Slot A athlon days with the bad AMD750 chipsets. I had an Asus K7M, and the 3Ware was not cooperating. After a few hours of fighting with it, I put it in my trusty celeron system with the BX chipset motherboard. It still didn't work, so then I knew I had a problem. This was unfortunate considering I was leaving for college on saturday, and this was thursday afternoon.
So I called 3Ware. I described the problem and had very little trouble (for once) convincing the tech support guy I wasn't some kind of moron. They believed me when I said I knew what I was doing when I put a system together, unlike most other companies. After an attempt to flash the card's BIOS to a good image, they said that the card most likely had a defective chip and would need replacement. They said they would gladly do it but their RMA system was down due to a server problem, and that calling hypermicro back would be the best option.
So I called Hypermicro, and they say they'll be happy to get me a new card, but there's no way they could get it to me overnight (as it's now late afternoon). I explain the problem of leaving for college and they say they'll be happy to ship it to me when I get there. I wasn't happy about that.
So I called 3Ware back, and explained what Hypermicro had said about being unable to replace the card before I leave. It's now about 8:30pm NY time, Hypermicro I believe is in central time, and 3Ware is on pacific time. The 3Ware support guy brings his manager on the phone (I didn't ask), and the manager says "Give me your information, and I'll have a card there for you. Send us your defective one back when you get a chance." So, 3Ware sends me a retail box card with the cables and all as a replacement for my OEM bare card, cross shipped, without an RMA number, and has someone take a car and speed to the FedEx hub in person to get it shipped.
10AM NY time the next morning the card was there.
The new card worked perfectly, and I was very happy with it. About 2 months after I bought it, 3Ware sends me an email. It basicly says the following: "We have implemented RAID5 support for your card in our new driver and firmware. You may download it as a free upgrade even though we had only planned to support RAID0, 1, and 10 on your product." How's that for support and driver updates?
Also, after the 6000 series was retired in favor of the 7000 and 7500s, 3Ware sends me another email that basicly says "We know your cards are considered obsolete and no longer made, but here's a free driver that adds > 137GB disk support to your ata-66 level controller that shouldn't officially support that standard." And at that point they'd basicly retired the products, yet they were still developing new drivers and firmware for their obsolete parts!
I now own a 7500-4 as my main controller, and I will likely be buying one of the 9000 series SATA-II controllers when they come out as an upgrade.
Oh and I've never missed my DPT (unfortunately since bought and ruined by Acraptec) with its hardware RAID5 and cache or my seagate cheetahs or any of my other SCSI hardware ever since.
Introducing the new Occam Fusion! Now with sqrt(-1) fewer blades!
3ware Storage Controller device driver for Linux v1.02.00.027
scsi1 : Found a 3ware Storage Controller at 0xdc01, IRQ: 15, P-chip: 5.7
You can easily make raid controller from a standard IDE controller, including one embedded in a mobo by simply soldering on resistor in place... do a google search for it.... tweak3d also has some instructions...
Words are only yours until someone else uses them...
I have now a 3ware escalade 7000-2, I ownz, very stable very fast (true hardware raid) very good linux support. I spent the spent the money on the 3ware after having problems with the onboard raid controller (soltek SL-75FRN-RL) and then bought amusingly enough one of the POS Sil680 based cards...these things are useless, corrupted my data despite numerous attempts to fix problem (new drivers bios update...etc the sil680 is just plain crap.) My G/F (who sometimes wishes I'd spend less on computers...finally told me to buy the 3ware card, cause I was getting so MAD at the SIL 680..she was worried I go nutz... I haven't yet done it but I'm totally going to go out in the woods with my shotgun and shoot the SIL 680. 3ware is very good.
actually I am happy to see you, however that is in fact a banana in my pocket.
I've used two other IDE mirroring controllers with great success although they are not in the
1. Arco IDE's DupliDisk2 product www.arcoide.com is a hardware based mirroring solution that is attached to your existing IDE controller. It requires no drivers and is platform independant. Great solid product I've been using for years.
2. 3ware Escalade 8500 SATA is what I'm currently using. It's something like $300 or $400, but it is also a hardware RAID solution and it supports hardware RAID 5, has excellent Linux driver support, and can support up to 12 drives on a single card. A very nice solution - as an added bonus, it is a 64 bit card.
That's all I've used, but again I highly recommend either of these two.
I'd rather be a conservative nutjob than a liberal with no nuts and no job.
I have one and I've tried several different configurations with it. However, I have been remarkably unhappy with it. First, it's a full-length card, I sure wish I had noticed that before I purchased it. Next, it's rather expensive, but does provide an onboard clock chip for RAID 5. The RAID 5 works reasonably well, and I'm sure I could pull a disk for a short time, but I wouldn't trust it for a long period of time. It was fine until...
If you have an issue, you're screwed. I had a drive lose some sectors, and everytime it would hick-up on those sectors, it would hardfreeze the box. When I changed it to not halt on error, it would just start beeping everytime it hit those sectors. but Win2k wouldn't mark those sectors as bad, because they weren't contiguous. Oh, and there are no tools on the Promise site to fix this particular problem. I just don't want to yank the disk because I'm afraid I'll lose everything else while I wait for the drive to come back.
Another thing is the lack of support. The website is pathetic. I tried to get debian installed, it didn't go so well. The driver they distribute is very badly written. It took me around 2 months to get it recompiled correctly, and even then, the sector errors freeze it (even with halt on error off).
All in all, I will never spend 300 dollars on a crappy IDE RAID controller and instead spend the money on RAID 1 something. Then I have full protection and can pull a single drive without worrying.
As soon as my full backup of the box is complete (another day or so to verify everything), I'm going to be yanking the card and reconfiguring. and not buying the high-end IDE controllers again... I wonder if serial ATA would be better.
Gonzo Granzeau
"Nothing the god of biomechanics wouldn't let you into heaven for.." -Roy Batty
Yeah, I tried one card based on the same chip. I attached two drives to the card, one which has data, and created a mirror set. The *#$#% thing spent 10+ hours trying to mirror a 60G HDD in BIOS, during which the PC can't be used, so I went to sleep and it was still going when I woke up. I cannot imagine spending 10+ hours each time I need to repair a mirror, so I scraped the PoS. Silicon Image seems to be big on SATA controllers, and I have one on my motherboard, but I have no drive to test it with.
I used highpoint hpt370 controllers and they worked well, but after I got a new MB which has suspend-to-RAM I found that highpoint's W2k driver hanged the PC if I used that. Not sure if it is just incompatibility or what, but I have also had IRQ sharing problems with the drivers so I kinda doubt their quality.
I now use a Promise controller in that PC and it has worked well so far.
My gripe with Promise & newer Highpoint BIOSes is that they erase the partition information if you change the mirror setups. I used to just add/delete a drive and make/destroy a mirror and use it as before (Highpoint's documentation seems to say this is okay). Now I have to backup the partition table to do it. It makes sense for splitting a mirror 'cos W2k gets confused, but not for making a mirror.
I have not really tried swapping a drive in a mirror which these controllers; with the above BIOS experience I am not very confident about not making a mistake and erasing the other drive in the process. OS-based solutions like Veritas are much better in this regard. Linux RAID also, but the user space utilities can use some polishing. As for W2k, it is only available on Server versions and it have to re-mirror after a BSoD each time, which is annoying.
I haven't tried hardware RAIDs but I am sure they are easier to manage than these cheap driver/BIOS based solutions.
A thought. Why doesn't someone make a RAID SATA controller in a hard drive physical format?
There's space at the edge, and physically overall.
Cables leave the edge and go to each drive (keeps cables short). While a singe cable goes to the MB SATA connector. Lather, rinse, repeat for the usual four connectors on a MB.
It frees up a PCI slot, and is easier to cool.
The 3-ware monitoring daemon will email you when a drive bites it! I got such an email a while back: put in a new drive after hours (no hot-swap) and told it to rebuild. It was done by the time I got home.
Jeff,
Here is some info:
When data is "striped" it is spread across more than one drive. If you remove one drive, the information on it is completely useless because only part of the information is on each drive. Striped is RAID 0.
When data is "mirrored", any data copied to one drive is copied identically to the other. Mirrored is RAID 1. If you remove one drive of a mirrored set made by a Promise controller, both drives have identical information. It is necessary to configure the one drive as a one-drive set, but after you have done that (from the BIOS setup of the controller card, before booting the computer), you can boot from and use either of the one drives normally.
It is possible to put four drives in a set that is both mirrored and striped. That is RAID 0,1. If you remove two of the drives that are one-half of the striped mirror, you can use each half of the mirror separately.
RAID 5 is 5 drives striped. Four are used for data, and the 5th is used for recovery information. If one of the drives fails, the system can continue without interruption. When the failed drive is replaced, the information on the other 4 drives is enough to replace the lost data and continue again.
When a mirrored set is read, the controller gets the data from the drive that has a head closest to the data. Reading is faster. When a mirrored set is written, the controller writes simultaneously to both drives. There is no change in write speed.
When a striped set is read or written, the controller positions the heads of both drives for fastest operation. Reading and writing is faster.
So, mirrored is safer, and faster on reads. Striped gives no data protection, but faster reads and writes. RAID 5 has both safety and speed. Since there are 5 drives, with 5 independent heads, reading randomly placed data is much faster.
RAID is "Redundant Array of Independent Drives", or "Redundant Array of Inexpensive Drives"
If you have other questions, post them as a reply to this message. Probably lots of people have the same questions.
Michael
The linux kernel list has not been too happy with Promise's support. In the latest Kernel Traffic, their support is starting to be reintroduced into the kernel proper, with some reservations:
"Elsewhere, Andre Hedrick was not so happy. After looking over the code, he complained that it was completely obfuscated, and added that he had severed all ties with Promise."
Be forewarned.
Promise is crap and more expensive it is the more crappy. I spent over $800 on a Promise Supertrack100 6 channel card. It was slow, slow, slow, and constantly corrupting data. Promise never bothered to make a new bois for it and the tech people blamed the motherboard (ASUS). I eventually did an "Office Space" on it - felt good. Now, 3-ware - there's a quality product. Solid.
I have used various Promise "RAID" Cards quite extensively ...
... simple, easy to install, and all around superior in every way; and my experiences with their tech support have been fantastic, in the rare occasion I have had to call ...
... save yourself some headaches!
In my understanding, the "RAID" stuff is actually done in the driver
The 3ware cards do the RAID functionality onboard; I have had MUCH better luck with them; they are Fast
Bottom line: the so-called "RAID" controllers (Promise, HighPoint) that are really nothing but IDE Controllers w/fancy drivers don't hold a candle to the 3ware cards; and I only paid about $15 or $20 more for my 3ware than I did for the last FastTrak I purchased
I have two Accusys ACS-7500 mirroring controllers. They work perfectly, appear as a single IDE device to the IDE bus and require no driver/BIOS/control software. It is a double-height 5.25" module which fits in a normal stacked pair of 5.25" bays.
As a test, I failed one 60G drive to force a rebuild, and it took about 8 hours.
Oh, you can migrate drive sizes upward, by failing and replacing one drive at a time. Next time the machine reboots(after doing both disks, of course), the controller appears as the new disk size and the partition/filesystem can be grown.
They even have a newer model (ACS-7630)which has three drive bays in the same double-5.25" form factor which can be configured as either RAID-0 or Raid-5.
"Employees at Fry's tell me that ECS has the poorest quality control of any of the motherboard manufacturers that they sell. To me, ECS motherboards seem to have surprisingly high quality. However, we have only tested three, and only one extensively."
LOL!!!....this guy obviously isn't a tech. I've seen my share of people like him who talk bullshit to impress their ignorant audience. So tiring.
I've had some experience with the Adaptec 1200A RAID controllers. I was planning on using them in about 5 workgroup file servers I support.
After I purchased them, I noticed that Adaptec's website says that Linux is not supported and they don't offer any drivers for this base-model card. However, if you investigate a little further, you find that the Adaptec cards use the Highpoint HPT372A chipset which IS supported under Linux. The drivers worked fine on a stock RH8 install and the GUI tools for RAID monitoring are exactly the same ones you get in Windows. You can download the drivers at Highpoint's website.
I emailed their support guys and told them they should at least provide a link to Highpoint's site so people that want Linux drivers for the card can be serviced. They replied and told me that Highpoint had provided them the stock Linux drivers for the chipset and they just haven't gotten around to actually customizing them, and that, in the mean time, they would just recommend people bump up to the next higher model RAID controller they sell.
Another caveat if you are planning on purchasing these cards is that they are only 5V PCI cards, not dual 3.3v and 5v or just 3.3v. Ergo, they don't work on some newer motherboards (like the Intel boards that come with the base-model Dell Poweredge 600SC server). I found this out the hard way and ended up going with 3Ware's Escalade cards which I've been happy with.
gee, Tandem has made mirrored disk controllers for oh, about 30 years.
guess what - they work great!!! all the time. no bullshit, no lost files.
can you say online backups? that's right, with a live database!
skip the PC junk, get a real fault-tolerant system.
You can get a NonStop system from HPQ.
(yes I worked at Tandem for 10 years.)
I rather like duplexing than mirroring (i.e. two controller cards) That way there is not a single point of failure, and cards are relatively inexpensive as pointed out.
I have a Promise SX6000, RAID5, WD200 HD's running under WinXP. Don't use the synchronize option! It _will_ corrupt your data. For example, I did a chkdsk (NTFS) to make sure the array is ok. Then I do a synchronization through the management program. After it is done, I run chkdsk again and get a bunch of random errors all over the hard drive. If you made the array your boot drive, the OS is toast. For WinXP, the sympton is a blank screen after the bootscreen animation is done. The Promise card does not work well with AMD related chipsets. It tends to crash under high loads (eg. dumping a large file across the network for backup) The management program can't seem to enable S.M.A.R.T. on the array.
Having installed RH9 on a new server, I wanted to make sure that the data disks were at least RAID mirrored.
The motherboard I bought has an onboard VIA 8237 Southbridge which has 2 S-IDE controllers (and an optional, Win2000+ only supported Software RAID bios). The stock RH9 does not recognize the VIA 8237, probably because its fairly recent. Newer kernels do, and with some patches to the IDE driver (adding the chips PCI ID's) the chip is now recognized ; the "old-style" IDE controllers work in DMA mode but it fails to find the two S-IDE controllers. End of story, as I had bought 2 S-IDE hard disks.
Second attempt was the Highpoint Rocket Raid S-IDE card (NT$ 3000) as it claimed to have Linux drivers. Standard the kernel will recognize the card as a Highpoint IDE controller.
You need to disable this (hde=ignore etc) and then compile and install your own RAID module. The code for the RAID module comes from the HighPoint website.
This I got to work, it recognizes the RAID array defined in the contollers BIOS. I setup partitions and an EXT-3 file system. But when I started copying files several GB of files to the new RAID array it crashed Linux hard ; and destroyed the filesystems.
No idea why; but as I didn't want to take any risks (and the Highpoint is Software Raid anyway); I disabled the Highpoint driver; and used the kernel IDE driver to access the Highpoint S-IDE controllers and installed the stock Linux software RAID solution. No problems, works charmingly.
Just my $.02 ; next time I have a larger budget I buy a decent hardware raid solution, not one of these half-baked software raid fakes.
This is an IDE RAID card that connects to the IDE port on your motherboard and then to TWO hard drives. Your motherboard (and thus OS) thinks it's talking to one drive (via the IDE cable), when it's actually mirroring to two.
Lots of alarming features, really everything you'd want, even background resyncing of new hard drives. The only problem is that they only have that background capability in Windows and Linux app forms, and some of my machines run other things, like QNX. No big deal, I just have to schedule a 30 minute outage to install a new hard drive and sync it up.
Of course if you want high-end performance and features, go with a hot swap SCSI cage. But for the rest of us, who just want to spend a couple hundred bucks securing our desktop data against [R-A-] I-nexpensive D-rive failure, DupliDisk is great.
I've posted about this product twice before, each time there was a RAID thread, but it was a while ago and I don't know how to dig up my old Slashdot comments.
One simple rule for its versus it's
There is a source drop available that has the kernel interface released as GPL. There is still a library that must be linked with, but this at least lets you recompile with whatever kernel you wish. I'm running the promise 20376 onboard the P4PE and kernel 2.4.21 on a debian (sarge) system just fine. I even boot directly from a 3 drive RAID0 array. Performance is great. I had tried hacking the pre-compiled version with limited success, but once I recompiled the driver for my kernel everything was great and very stable. :)
You'll want to do a search in google for ft-par_v1.00.0.15
Then just recompile and you'll be on your way
Regards,
Paul J. Rider.
The array switches to Disabled state at random, with no apparent cause. This happened twice yesterday, but not at all for the previous 8 days. I contacted Highpoint support and their advice was (1) make sure the BIOS version matches, and (2) check cables and connectors.
I'm using a beta version of their opensource driver which is not linked from their webpage. I put it online at hpt374-opensource-v2.01-0718.tgz for any others who need v2 driver source rather than v1.
I'm running 2.4.21 with LVM. I did have the root filesystem on LVM under the Highpoint controller but had problems with extending it. So it is now under the inbuilt IDE controller, using software RAID and no LVM. I find my box is less fiddly to configure if the root filesystem is not LVM.
When the drivers set the array state to "Disabled" no more disk I/O is possible to those devices. A reboot is required. Strangely, the devices have never required resyncing after this problem.
At the subsequent reboot linux usually spends a few seconds replaying the filesystem journal (they're all EXT3 filesystems).
I have a theory that the Highpoint card (or driver) does not like sharing interrupts. Since the problems yesterday I have fiddled with the IRQ assignment and one problem that I had noticed (occasional ghost streaks on the CRT display, signifying contention on the video memory during a raster trace interval) has apparently disappeared. It may be that the Highpoint controller was sharing an interrupt with the display controller, making one of them not very happy. Thus, I'm cautiously optimistic that my problems may be cured.
In summary I can't really recommend the highpoint card for stable use. It's great to get a lot of IDE channels (e.g. using the quad-channel ATA card, and putting 2 devices per channel) and SATA is great for cabling drives, but there are obviously some driver issues which need to be worked out first.
We're having what might be a slightly unusual problem for a mirroring IDE RAID card: excessive numbers of hard drive failures.
A bit of background: We have 550 sites with Dell Optiplex GX240s. Most installations have been in place for around a year and a half now. They are configured identically with Promise FastTrak100 TX2 cards, all of which are in the same PCI slot (except for a few test machines which were used to verify that the PCI slot was not the issue... these test machines had the same rate of failure as the main configuration). The card is on its own IRQ line, even though IRQs shouldn't matter anymore. When we purchased the systems from Dell, we received two drives per unit and installed a postfactory Promise RAID card and modem. The software is Windows 2000 Server with SQL Server 2000.
Since Day 1, we've noticed that a far greater than normal percentage of drives fail. Of the 550 systems, two or three a week will suffer a hard drive failure. When we received the systems, the initial mixture of drives was about 70% Maxtor, 20% Western Digital, and 10% Seagate. 95% of failures are Maxtor drives, and 95% of failures are of the drive attached to the primary controller.
Very recently, we did some testing. We took a known good hard drive and plugged it into a controller that had toasted a drive. We ran some disk churn creating programs and the previously good drive failed. We ran the hdd manufacturer's diagnostics on another system before and after, so we know it was good before and bad after.
Now, my question... has anyone ever seen this before? Is there something wrong with a controller or with a driver or something? Anybody have any troubleshooting ideas? We've been banging our heads against this for a year now and can't come up with any good explanation for the results we're seeing. And calling dell tech support to return 20 drives is not something anyone wants to do because they make you run their "diagnostics" -- a five minute process that tells you the drive that fails the manufacturer's diags with severe failure errors and such is really a good drive.
Is this trolling?
Nothing nibbling here.
Skot Nelson music is my saviour / i was maimed by rock and roll
A large number of techs thankfully never have to deal with gameboyz clone mobo junk.
Tech-Report did a review of various RAID controllers last year.
In addition, VIA's KT600 provides Serial-ATA RAID 1 (mirroring) capability through the southbridge. There are also a large number of nforce2 boards that use Silicon Images Sil 3112A controller.
The dominant reason for mirroring is data availability. In linux, the OS is perfectly capable of doing this perfectly well. So we need a reason to have a _hardware_ raid card. There are three that I can think of:
;) Share partitions over NFS, or network protocol dejour. Gigabit ethernet should exceed the throughput you were actually getting from those cards. (ATA 100 and 100mb ethernet are approximately the same speed...)
1. Additional IDE channels.
2. No use of main CPU (although this apparently is NOT true of many of the listed cards, it would be a good reason)
3. Cache. A real raid controller caches the last 64MB at least of transactions, which especially avoids lots of seek-time waiting, hugely reducing disk thrashing. More is better. In linux, your main memory already does this for you, but it uses the main CPU to do it.
4. Battery backed cache. A real raid controller has a battery backup for the cache, solving the problem of worrying about your unsynced (to the drive) data in a crash without doing those unoptimized writes. This only matters for writes.
My ultimate mirroring controller is *drumroll* another linux box as a NAS. Put 1.5 GB of RAM in it, most of which will automatically be used as cache. Don't boot into X
For added fun, install other file services, and make it really work as a NAS.
To optimize for cost, use whatever hardware you have laying around. Anything with PCI should work adequately. Machines with at least UDMA will be more responsive (generally PII +) and for bonus cheapness put 4 disks on it. (1 of each mirrorset on each chain) So the cost for many people is free.
For real performance use a MB that supports two channels of the ATA speed the disks really support. And buy an IDE controller (or a scsi controller) if you need more disks than your channels support well. Practically speaking, ATA66 is still almost enough for most common drives right now.
My mediocre answer to battery backed cache is a UPS.
The next complaint is usually about being able to mirror the boot drive if your mirror is in the NAS. My answer to this is to go ahead and softmirror the boot drive in the workstation. But try to have very little written to that drive that will be waited for, and very little read from after boot. Even better is to netboot from a CD or floppy you have many copies of.
Looking for freelance Actionscript (Flash/Flex) or ColdFusion work and/or freelance developers. Email me, put Slashdot
Fastrak Source
/proc/scsi/FastTrak or something like that if I remember correctly...
I've been using this to support our TX2000 Pro cards under Debian, seeing as though the pre-compiled drivers are for winblows/Redhate/Suse...
Worked fine for me. Not huge on stats, but you can cat
I've been thinking about RAID in my home box lately too. I like spending more for a nice system but not to the extreme 'god box' levels that some do. For example, I went with a dual athlon 2000 setup recently for only about $800 (mb+2cpu+mem+ps) but while I'm very happy with this setup I feel like I've shorted myself on the hard drive setup.
The plan I've been mulling about is to have SATA RAID 0 with two 60 Gig drives and then a 120 Gig drive that I rsync up with nightly so that I have a relatively easy way of recovering things if they fail. This is quite a bit more money than the single drive I have now but it seams reasonable compared to going SCSI or 10Krpm or a 4 drive setup with a card that can handle it.
Like I said, I'm just mulling it over and this is the perfect place to put it out for comment so... what do you think? Has anybody else tried this and been happy with the results?
Davo -- Free speech, free software, AND free beer.
Has anyone used such a product?
Would something like this be good in a desktop/workstation?
How about an ATA version? Does one exist?
I work as a sysadmin, and I am also a developer for Gentoo Linux.
Thru my career, i've used a lot of different cards.
My recommendation is solidly on the 3ware for truely solid hardware. It's is practically the only pure hardware RAID solution available. It shows up as a SCSI device to the system, and your system just needs to know that, not that 2 IDE devices are connected.
I've had horrible times trying to make the promise cards, esp the really new ones found on Tyan motherboards work, there is practically no support at all for them in linux (binary driver is no good since you can't even see the hard drives to install onto them).
The highpoint units are very good budget units. They have gone thru plenty of testing, and ship a good product, including some very nice UDMA/133 cables.
I have been in direct contact with the highpoint driver development team for the last several months helping them iron out bugs with their v2.10 linux driver, which finally contains RAID5 support now. I specificly helped them clean it up so that they could release the majority of the source code (there is a small binary part that appears to contain RAID checksumming routines and some properiatary card interface calls). There is also a largish binary only gui config tool, that I haven't personally used as they can't compile a version for my non-standard glibc setup.
If you have the money to spend, and just need mirroring, I'd put my full recommendation behind the 3ware card.
However, if you want a more budget option, or want to go further than 2 disk mirror/stripe and don't have a PCI-X slot, the highpoint is a good choice.
if you have PCI-X, go right and buy the higher 3ware models, you will not be sorry whatsoever.
I am currently running 8x 250gb maxtor drives on a highpoint controller at home for my personal storage unit, as RAID5, providing roughly 1.75tb of usable space.
ICQ# : 30269588
"I used to be an idealist, but I got mugged by reality."
I have the adaptec 2400. Had it in raid 5, but didn't have a hot spare installed. When one drive failed, I ran out, bought another drive of equal size, and installed it, then tried to rebuild the array. Didn't work because apparently, oem drives are ever so slightly smaller than boxed drives (of the same size, 60 gig), so the array refused to build because it said the new drive is smaller than the drives currently in the array.
Great.
Didn't have time to return to the store before it closed. Weekend. So I ended up transferring the data (was less than 60 gig between os and data at the time, even though the raid 5 array was 120 gig capacity) to the new drive on a different controller, then rebuilt the two good drives from the raid 5 into a raid 1 and reinstalled everything.
To make a long story short, eventually ran out of room on both drives, one on a separate controller, one in a busted raid 1 array on the adaptec card. So now I have no redundancy currently. And the drive in the busted raid 1 array is a deathstar...
So, I've been trying to figure out what my options are. I don't want to buy another adaptec card for a second raid array. And I don't want to buy a large drive to move data to temporarily from the two 60 gig drives, because I'll be wasting money on the large drive since I can't use it in the raid 5 array when I'm finished transferring data.
Since I also have experience with failed drives, I won't be using raid 1, and returning MY DATA to the hd manufacturers for warranty service. My only option for protecting MY DATA with a failed drive is raid 5. So I need 3 drives, plus a hot spare this time around.
I have the adaptec 2400A. But the adaptec has 4 channels. While that allows raid 5 plus a hot spare, it also only allows two data drives plus one drive for parity. But with raid 5, you can also add another drive, for three data drives plus the parity drive. With a hot spare, this would require 5 controllers, not 4.
I've looked at 3ware's site in the past. While I have seen that they do have cards with more than 4 controllers, I haven't seen any mention of adding a 5th drive, for 3 data, 1 parity, 1 hot spare. I believe I've even downloaded user manuals on some of their drives, but haven't found what I'm looking for.
Since you seem to have some knowledge/experience with 3ware, do you know if this is possible?
Raid 5 is great. Until you run out of room. Then what? I never thought I'd run out of room with 120 gig of raid 5 storage (plus 60 gig parity). But with partitioning, multiple operating system booting, and not having that hot spare installed from the beginning to ensure it would build...I ran out of luck.
OK I already blew 3 moderator points on this thread but it's time to negate them with this post. :-( Bye-bye points.
The long and short of it is that all of these cards are not very good. As has been mentioned before, these are all really plain-old IDE controllers with proprietary software-based RAID built into the drivers, and an inflated price for that package. In other words you're better off using plain IDE controllers and using Linux's built-in software RAID.
Why? Linux's software RAID is faster. It outpaces all of the binary-only software drivers that companies like Promise and Highpoint provide, and is better than the reverse-engineered drivers that are in the kernel.
I know -- I have a motherboard with a 4-channel Highpoint 374 controller on it. Getting it to work properly under Linux has been a nightmare -- and the end result is still slow. I finally broke down and used it as a plain IDE controller -- and it works passably in that regard, but it still has tons of stuff not implemented in the kernel (and the binary-only drivers are even worse -- they allow access to the disk blocks and almost nothing else -- so forget about using any advanced features of your drive).
If you really want REAL hardware ATA RAID, the best choice is 3ware's Escalade series. Not only does the support in the kernel kick ass, 3ware's support is excellent. Barring some major changes in the marketplace, I will never get ATA RAID cards from any other source.
That said, Promise and Adaptec do apparently make at least partially-hardware-based ATA RAID controllers, but I don't know how good the support under Linux is... Most importantly these are their expensive controllers, not their cheap cards -- they are in the same ballpark as 3ware, so there's no reason not to choose 3ware.
Finally, don't expect miracles from ATA RAID in terms of speed. While it's great for the added redundacy, speed does not go up linearly. Basically what happens is that while your transfer time goes down, your seek time stays the same. Assuming an arbitrarily fast bus (think something on the order of 133Mhz/64-bit PCI-X, not lame 33Mhz/32-bit PCI that is in most consumer desktop motherboards), as you add more drives eventually the vast majority of time your array is seeking rather than transferring data, and seek time becomes the bottleneck.
This is why if you are serious about RAID for speed (rather than just for rendundancy) you really need to go to a 15Krpm SCSI array. 15krpm drives have slower media transfer rates due to lower data density -- but you can always improve that by adding more drives to the array. Your seek time never improves. This is why a WD Raptor can compete in the same ballpark as a 15Krpm Cheetah, but when put in an array the array of Cheetahs THUMPS the array of Raptors.
Also I recommend a SCSI array because of command queueing and reordering. Supposedly Serial ATA has support for reordering but I don't think anyone has actually implmemented it -- so far all Serial ATA implementations are just plain old ATA-style traffic over a serial line. Under heavy disk usage access over multiple files, SCSI is smart enough to spend way less time seeking than ATA. (As I understand it the 2.6 kernel tries to do reordering in software to try to make ATA drives perform better in this regard. I will be interested in seeing how well this works.... But really the kernel does not have better knowledge about how the drive is seeking than the hardware, so I suspect it will not be as effective.)
1. Promise don't work well under Linux at all.
2. I've seen silent data corruption on some machines using Promise, both running Windows and Linux.
This leads me to the general advice to never buy Promise mirroring controllers. It's your decision of course, but please do yourself these favors:
1. Keep backups
2. Keep backups
Also, Anandtech and Tom's have had some reviews in the past.
Personally, I've had bad experiences with my onboard Promise in Win2k... worked fine with 40GB IBM, but when I got a 60GB IBM, vague errors started appearing in event viewer. Promise & Gigabyte did not answer any of my emails. IBM said they normally like Promise, but had no suggestions. I gave up and plugged back in to the Intel controller, and haven't had a single problem since then (over 2 years now).
I've also seen vague errors on Linux when using Promise controllers, and they still don't have an open source driver.
Nothing to see here; Move along.
..they do not handle drive failure well at all.
I had to deal with a system that would mysteriously lose its own drives after it attempted to boot. It turns out the Promise FastTrak controller (integrated onto the motherboard in this case, a early Thunderbird ASUS board) couldn't handle a read error on one of the drives, and subsequently took itself completely offline.
The worst part? Absolutely no indication whatsoever of the problem. The statistics available from the Promise interface said everything was working just fine.
If you want reliability, these cheap options don't cut it.
I've been reasonably happy with the performance and reliability of the storage subsystem using it, (although being a paranoid sysadmin everything on the RAID array is also held on another server as well as being backed up via a third!). Personally while I am happy with the Promise card I would have prefered to go with a 3ware card as they seem to have a better specification and better driver support but the availability is not great in the UK, (and of course the cost ... these are personal boxes not commercial ones. :)
I have also been using the onboard promise controller on several older (ASUS A7V133) motherboards but these are jumpered not to use RAID support (just to act as additional normal IDE ports) and then use Linux Software Raid on top as I've found it to be more reliable and produced higher performance.
t
- Promise FastTrack100 TX2 (PDC20270)
- Adaptek 1200A (HPT370A
chipset),
- Iwill (HPT368 chipset)
- HighPoint Rocket 133 (HPT302 chip: didn't
work at all) some months ago
- and PDC20276 built onto the Asus P4B533-E motherboard, as well
as the
- various CMD64x, including the CMD649 and the
- Silicon Image 680
I returned to the Si680, but still keep getting:And worse, I get random lockups where the keyboard lights do not respond, Alt-Sysrq-s has no effect, and nothing in the log indicates anything at the point of lockup.So I'm off to buy a 3ware to keep my 6 hard disks going.
It's going to be a nuisance to back everything up (203GB) and then restore, but a man's gotta do what a man's gotta do. Unless you have any other suggestions? All are most welcome.
If you want to use RAID 1 you want the sequrity if one drive goes to bed... in other words, your data is important.
/Christian
Only one solution, you go and buy a REAL server from a branded company (HP, IBM, etc) with SCSI disks and a real hardware RAID-controller! Sure, cost more but you get what you pay for....
If you still go IDE you still should buy the whole server from a branded company.
Again, if you want sequrity for your data, money is of secondary concern or else your data isn't that important....
I didn't know that the Promise controller is a "semi-software" RAID card. What is the shortcoming of using such cards? What extra value do the 3ware cards provide? Why are software dynamic drives better?
Promise controllers provide a way to make a working backup copy of a Windows XP boot drive. Since Microsoft provides NO way to do that, some method must be found, and Promise controllers provide one of the methods.
That wonderful firmware update isn't unusual for them, either. When I bought it, my 8500-8 SATA RAID card (*drool*) didn't support hot addition and removal of entire arrays. One firmware update just quietly mentioned in the release notes that they'd implemented hot addition and removal of RAID arrays. Wow.
/really crap/. Anyway, once I got the card replaced and found someone at 3ware who had a BRAIN ( and was absolutely brilliantly helpful actually) everything was peachy. I'm very happy overall, though I wish they had better web support and card quality control.
My first 3ware card was a horrible dud, and actually caused me a lot of grief. My local dealer was really bad, and as I'm in Australia 3ware wasn't able to help me directly. Oh, and their web support was
As such, there are three ways of getting them working with Linux:
Use the Promise RAID driver
Use the Linux ATARAID driver
Use the Linux md RAID implementation
The first of those is handy if you're dual-booting with a RAIDed Windows installation and want everything to work. The disadvantage is that you'll be limited to certain kernel versions as other posters have already noted.
The second option also allows interoperability with other OSs RAIDed on the same drives, but because the drivers aren't written by Promise, there may be some gotchas. The advantage is that ATARAID comes as standard with all Linux kernels, so you'll never be forced to lag behind through lack of driver availability.
The third option is probably the most stable and convenient, as long as you don't require another OS to use the Promise RAID setup (md has not been ported to Windows, as far as I know!)
I chose the md option as although I do dual-boot with Windows, I don't have any important information stored there. Note also, that because all the smarts are done in software, there's no inherent performance overhead in using md over Promise's driver.
--
I know it's not supposed to be a low-end RAID-card (however, it IS :>) like the mentioned ones, but probably you also considered this one. Don't even think about it. I shared my experiences on StorageReview, click here to read it.
Free your mind!
I'm using the Promise FastTrak100 card under XP with dual 160Gb Maxtor disks and its working a treat so far.
Email me oli-spam-my-arse-off@ojms.org.uk if you want to know more.
As it happens last week I got a machine with a Promise Fasttrak TX 2000 card in it. I wanted to mirror two disks.
Unfortunately, I discovered thatthe promise drivers are proprietary and that the drivers contain all the smarts - this is *SOFTWARE RAID*, just with a BIOS control screen to configure it.
I spent a couple of days trying to get this working on Debian - and I'm no linux slouch, but I was beat. In the end I got it to recognise the controller and used the kernel's standard disk mirroring - which some people claim gives better performance than the promise controllers anyhow.
Basically, these cards are pointless if you are a linux user. The only reason I can see to use them is if you want the same drives used as mirrors for both Windows and Linux on a dual-boot system. Otherwise, leave well alone.
Does anyone have experience with ICP Vortex Raid controllers? I used a SCSI model about 5 years ago and seem to recall that it was one of the better controllers I had used. Looking at their site I see that they now have a SATA controller that supposedly supports Linux, FreeBSD, Unixware, SCO, Netware and Windoze Nt2kXp.
Do not buy these cheap mirroring controllers. Use software raid from your operating system instead. It's more stable, easier to recover, and runs at the same speed. As others have pointed out, these controllers are doing everything in software anyway, so you may as well hand the responsibility for the data over to the more robust and mature OS drivers and save yourself the cash.
If you want real RAID, spend the money. As many have pointed out, 3Ware owns this market. A 2-port 3Ware card is $120, and they offer 4, 8, and 12 port models. They do standard ATA or serial ATA. As of now they have a limitation of 2TB max per array, so if you have more than 2TB of space you'll need to divide it into two arrays to use it all. This limitation is going to be removed in the future.
Another thing you really must not overlook is proper cooling for these systems. Even two IDE disks in close proximity get pretty hot. Not as hot as their SCSI anscestors, of course, but the heat adds up, especially on busy arrays. I highly reccomend that if you are serious about protecting your drive's health you invest in some kind of hotswap cage. These have dropped drastically in price. Look for Cremax ICYDock bays. They are cheap, you can get a temperature monitored, fan cooled, hot swap capable mounting bay with alarms and a lock that will mount 5 drives into 3 5.25' bays for $150. That's a steal. They offer a 3 drives in 2 5.25' bays option for $110. Check out www.scsi4me.com, it's a good site with a good selection.
I've done break tests on numerous 3Ware arrays, and had them rebuild perfectly every time. They have a much better track record than SCSI controllers. I've also benchmarked them against top of the line SCSI controllers. 3Ware decimates SCSI in writes, and ties or lags just a tick behind in reads. SCSI is still the king of access time, however... but for the 3x increase in price over ATA I can live with waiting that extra 6ms.
There ARE some things ATA raids lack, however...
So far, I've never seen any IDE/ATA raid controller that allows you to *grow* an existing RAID5 array by adding another disk to it on the fly. You always have to delete the array, then recreate it using the extra disk. I know a lot of the more expensive SCSI cards could dynamically grow their arrays. I asked 3Ware about this when I stopped by their booth at Linuxworld and they did say it is something they plan to offer in future products.
Some of the very expensive SCSI array cards would also stripe the free space of your online spare. This was handy because it used all the disks on your controller at once, giving you the best speed. It also let you know in a hurry if your online spare was a lemon, since all disks were in use. If you had a disk failure, it would then regenerate the missing data into the free space striped across all disks, and you would end up with a perfectly healthy RAID5 array and one dead disk. You could then lose your standard single disk from the RAID5 and still have your data even though the array was degraded. I mentioned this to the 3Ware guys as well and I was surprised to have them tell me they had never heard of it before. Again, this was a common feature on the high end Compaq array controllers. I'd love to see this one make its way into 3Ware's products.
Hell is being intelligent in a world full of idiots.
I also have a small 3Ware card in another computer. I have a 7000-2 in a desktop machine. I haven't had any problems with it. I prefer 3Ware cards frankly. I haven't tested their Linux support yet but I suspect it will be just fine.
The onboard Promise FastTrack133 card on my Asus A7V333 board sucks the big one. If I so much as have a drive connected to the RAID headers, the CPU load spikes once a minute regular as clockwork at roughly 30% above the current load. This makes things like MPEG playback, MP3 playing, and games stall for a second or two. You don't have have to have the drives formatted and mounted. They don't even have to be configured for use in the Promise BIOS controller. They just have to be connected. It's a joke.
Those are my experiences. You get what you pay for apparently.
One is two sets of Raid 5 mirroring each other, a derivative of the more common Raid 5 in two Raid 0 stripes (used for speed, usually with two external hardware Raid 5 enclosures and Raid 0 by a PCI card or software. Note this mode is natively supported by Apple's Xserve RAID).
The other is the opposite, with each drive in a Raid 5 configuration mirrored. This is only used by very large organisations, but if you have the money, it is the best of the best! If a drive dies there is no slow down at all. Two different sets, no slow down. A whole mirrored set, a slow down, but the network does not go down! Many disks need to go down to cripple this setup.
Mr Smyth,
Wanted to talk about opportunities with a few of our sites. 506-855-2991 x111
georged@traffixinc.com
I have a Zip 100 Disk that won't work without the cable because of the interference of all the other drive cables running in parrallel. BIOS will detect it as "Z@#$1@#$@" and it obviously doesn't work.
Also, I've had a Seagate that wasn't very tolerant of interference. It was getting CRC errors about 10+/s without a good shielded cable.
I don't know what tests you've seen, but the tester must have been an idiot. Run four IDE devices (at different speeds) in your box with the crappy standard ATA66 cables, in parralel, and then get out the smart utility from Linux and start transferring from all the devices. I bet you'll find more than CRC errors.
I have run into the problem myself, and I'm sure a many others have too, so you can bite me.
Just think of it theoretically as well. Why do you think there is a standard Cat5STP (shielded)? Why is RG6 shielded and required for satellite systems to work? Cables are lossy at high frequencies without proper shielding. You're probably one of those people that make ethernet cables by sorting the colors alphabeticaly (or by the colors of the rainbow, or make an acronym out of the names of the colors).
I have better things to do with my time than read the dictionary or play scrabble, consequently, I have very little respect or time for you Grammar Nazis that don't yet comprehend logic. While you're learning to spell some arbitrary word, I'm testing ATA-133 cables.
Now would be a good time for you to buy a book on making logic arguements and keeping irrelevant issues seperate.
If you don't want to take my advice on anything, then don't. Just don't come here posting non-sense without evidence.
Karma Clown
AIX ripping off Windows? What? I'd say your an underling for some real Administrator, or perhaps your network is mainly Windows based and your the sole Linux champion. Please your insulting /.
Reading of data is considerably faster since the controller reads the data from the drive that has a head closest to the data.
Two disks in a RAID 1 mirrored pair will not improve read performance over a single disk of the same type. The disk controller has no real information about the drive's physical geometry. So a RAID 1 controller cannot accurately determine which disk's heads are closer to the next data to be read.
There was a time when the hard disk controller knew about the actual physical layout of the disk - how many physical heads, how many cylinders, how many sectors, and so on.
But for many years, hard drives have obscured their physical characteristics from the disk controller. This was done to overcome BIOS and OS limitations on the number of addressable cylinders. You've probably heard of this development - Logical Block Addressing.
From http://www.wikipedia.org/wiki/Hard_drive#Operating _system_use_of_hard_disks:
SCSI drives, however, have always used LBA addressing, which describes the disk as a linear, sequentially-numbered set of blocks. SCSI mode page commands can be used to get the physical specifications of the disk, but this is not used to read or write data; this is an artifact of the early days of SCSI, when a disk attached to a SCSI bus could just as well be an ST-506 or ESDI drive attached through a bridge (and therefore having a CHS configuration that was subject to change) as it could a native SCSI device. Because PCs use CHS addressing internally, the BIOS code on PC SCSI host adapters does CHS-to-LBA translation, and provides a set of CHS drive parameters that tries to match the total number of LBA blocks as closely as possible.
ATA drives can either use their native CHS parameters (mainly on early drives; most hard drives made since the early 1990s use multiple-zone recording, and thus don't have a set number of sectors per track), use a "translated" CHS profile (similar to what SCSI host adapters provide), or run in ATA LBA mode, as specified by ATA-2; LBA mode generally has to be requested explicitly by the host computer. ATA drives larger than 8 GB are always accessed by LBA, due to the 8GB limit described above.
Given these conditions, a controller (SCSI or ATA or SATA) is presented with sectors as a linear sequence. So when the controller receives the request for data at sector 20, it needs to determine "Is the distance from sector 10 to sector 20 further than 110 to 20?" The controller cannot really answer that question.
-Don
The big thing to look out for are 'RAID cards' that are actually just regular old IDE controllers with RAID features implemented in the driver. You didn't specify OS, so YMMV
In Linux, the only thing these soft RAID cards will give you that the md driver won't is a compatibility nightmare if you ever have to replace the card. Unless you're going to spend the bucks for a real hardware RAID card, just get something linux can see as a regular old IDE card and use the md driver for the RAID. It'll be faster (most of the vendor provided soft raid drivers are crap) and you won't have to use the exact same card when the time comes for a replacement.
In general, the promise chips can be used like that (though you may need to add it's PCI ID to the regular Promis driver's list to get it recognized), and if you disable the option ROM in BIOS, you don't have the long delay while the firmware trys to get it's act together.
I've been using Promise controllers for a while now, with varied results based on method of implementation. I can't say that I'm extremely happy with them, but I've seen a lot of old info posted so, for the record, here's the deal:
/dev/hd? by default, and using them is guaranteed to destroy your array. I've also had some random kernel panics, and that doesn't fly for an HA system...
Promise has not updated their literature in a long time. They've release some source for the driver so you can run a custom kernel and the distro of your choice. You're not dependent on Red Hat or SuSE, but the binaries are still available.
Building the driver is not difficult, but you do have to edit the Makefile to fit your system. They don't provide much help, but I think most people running RAID in linux on something other than Red Hat or SuSE are going to be comfortable enough to do it.
Booting from these things suck. If you don't mind building the driver and making an initrd image to preload it, you're on your way. You should also be very comfortable with your bootloader and go over some of the linux IDE and SCSI boot parameters. If you're all of the above, it's still not a good time...
I've had some problems with the kernel support for the 20268R. Works on and off it seems. When it doesn't you have to rebuild the array and hope you didn't loose all of your data. Also, you still have access to both discs via
Now for my suggestion:
Since the Promise proprietary driver works, use it. Don't bother trying to boot off the array. Get a small disc for the system to boot from and run everything else off the array. If you want, get two small discs, use the card to make a mirror, and if the boot drive dies, plug in the other. That's the config I finally settled on, and I haven't had any downtime yet. If I could go back, I'd probably choose 3Ware, but everything is working fine so I won't complain...
ascii art
The disadvantages layed out for Promise are pretty weak, mostly insignificant. I don't care about their Web site or their literatures; however, the point about Windows XP would be important.
I use Promise cards and they have worked very well under FreeBSd, which has native support for their cards in the ata(4) drive. I can also use atacontrol to see the status of the RAID, as well as see messages in dmesg about it.
I would totally recommended these cards for FreeBSD.
This is all very interesting information. Thanks for bringing it up!
On a related note, how does RAID 0+1 compare to RAID 5? I've got an old Mylex card that has both options (among others).
-Rich
It's very easy to install knoppix on an ATARAID controller... I did this on a HPT370 on my VP6.
= 17&mid=78&rid=
These are the changes made to the knx-hddinstall (v0.38) script and to the lilo.conf that makes it possible.
http://www.northern.ca/forums/index.php?t=tree&th
-ironstorm
I noticed you as as a newcomer to my freaks list (you marked me as a foe). I can usually figure out why most people who mark me as foes, but I just read through some of your comments and it's not obvious to me where you might disagree with me. I usually do tit-for-tat when it comes to that sort of thing, but in this case it doesn't make sense.
So, why am I your foe?
If tits were wings it'd be flying around.
on the plus side, their support staff was very responsive and even made a follow up call to see if I was happy after I made a lot of noise over the card's lack of support for drives >130 GBs
[Fuck Beta]
o0t!
Greetings!
This is Joel Franklin. In late 1999 we were briefly colleagues at 21CS and then were lay-off comiserators. It's been a while since we communicated. I've been checking your slashdot comments periodically in hopes of replying to one quickly enough to be noticed by you (assuming you go back to check for responses to your comments). This seems vaguely creepy and stalker-ish but after losing touch I didn't know how else to contact you. I recalled your slashdot nickname and web address but no longer have an email address for you.
Anyway - I'd be delighted to hear from you. Drop me an email when you have a chance. My email address is my full name in lowercase with no spaces + at symbol + "kc.rr.com"
P.S. I was RIGHT about the Oracle in Matrix I!
And here, ladies and gentlemen, we find one of those rare areas where "proprietary UNIX" truly excels.
The reason you buy a Sequent isn't because of ptx/Dynix (...of that I'm sure). You don't buy an RS/6000 for AIX (well, not entirely). Same goes for Data General, HP, AT&T (nee NCR), Sun, Encore, etc etc.
These boxes have storage management that ranges from just adequate (sun "software raid" == solstice anyone? :-) to mind-boggling (Data General running CLARiiON). LVM, duplexed controllers, redundant Host Bus Adapters, MPIO (Multi-Path I/O), and of course redundant power supplies (not "N+1" which is not the same...). I haven't found a way to have all these together and eat my cake too, when it comes to X86/PC hardware.
In short, you simply can't buy "COTS" hardware to do the same job - though IBM (Shark etc), EMC (Symmetrix and now CLARiiON), and StorageTek will sell you "turn-key" solutions that will have your wallet dripping blood - but you'll never lose a drop of data.
Just my $0.02. Linux rocks, but it has a way to go before it can support "commodity enterprise hardware"; you see, that's a contradiction in terms and a dream that may never become fully realized --- the trend with enterprise hardware is away from commoditization, while more and more often, some or all of the software is free (and supported). The margin has to be there, whether it's hardware, software, or services, for vendors to be willing to play in this space.
I'm not speaking of least-common denominator hardware, here, but of high-availability platforms with 7x24 support and 4-hour response times, guaranteed by the seller/manufacturer. There has already been great progress in devising low-ball/open-source approaches to availability across the enterprise, and that will continue, but there will always be a market at the high-end.
Redundancy is good; triple redundancy is twice as good! - Me.
"The plan I've been mulling about is to have SATA RAID 0 with two 60 Gig drives and then a 120 Gig drive that I rsync up with nightly so that I have a relatively easy way of recovering things if they fail."
In short, yes. Some time ago, I would unmount a drive and use dd to mirror it daily, though it required an identical drive.
Rsync is your friend and note that you can separately synchronize your mail spool more frequently then your binaries. But... this does not account for human error or malice: my hourly cron job diligently mirrored my one hack before I could deactivate it. Doh! Solution: hourly, daily and weekely mirrors (rsyncs), allowing you to rollback. To be really safe, mirror your two 60GB's and have one rollback on your 120GB.
MD.
In about 1998 I spent the dough to get a DPT controller (about $300 used on ebay at the time, since they were over $1000 new) and 6 4-gig Seagate wide-SCSI drives (5 to use, one as a spare). I mounted the 5 drives vertically in the front of a rackmount case, with a little space between them, and a 6-inch muffin fan above them to suck hot air upwards and out the side. Well... those drives sure ran hot. A mere few weeks later one of the drives quit. I forgot some details about what happened next, but I wasn't impressed with DPT's ability to recover. Basically the system was useless until I had installed the spare drive and rebooted; and it kept going downhill from there. I had other problems with getting it to recognize all of the drives part of the time too, as if it was some kind of termination issue, or electrical interference getting into the SCSI bus, or something like that. But it was a good cable and had proper termination.
/home from one to the other every night. I don't feel completely comfortable that data loss is impossible, but at least I can check that the backups are still happening now and then, whereas with the RAID I eventually lost everything and had to restore from the previous hard drive which I still had lying around. When I get a DVD burner I could probably get most of /home to fit on one disc again.
And those controllers had such a good Linux reputation at the time, because they were the first RAID controller mfgr to cooperate with getting a Linux driver developed.
So now I have a couple 100-gig IDE drives and have a cron job do a backup of
At home I have 2 golf-cart batteries in series and a little battery charger keeping them topped off, and some 12-volt wiring to a couple places in the house, including a 12-volt fluorescent light in the kitchen. It can be useful when the power goes out. But several companies now make Via Eden-based systems which can run from a 12V input; so I'm thinking pretty soon I want to build some new servers with these to replace the AC-powered ones, and run them from the batteries. That way they will be completely isolated from power surges (except for the charger, but a transformer is involved there, which provides some isolation) and ought to be able to run for years at a time uninterrupted. As well as being fanless and quiet (but then I need cool-running, quiet hard drives). Preventing power surges might even make hard drives last longer.
/usr, for example, because it can be replaced, and usually it's better to get the latest versions of everything anyway. (But I do back up /usr/local, /var, /home and /etc.)
Backups of selected directories could be done over the network from one to the other. I don't bother backing up
It's based on the HPT370 chipset, which is supported as far as I know. I've used my 1200A successfully in Mandrake 8 and 9, which I am going to assume makes it compatible with any recent linux release.
Maybe I can shed some light on why some people have trouble with K7S5A motherboards, and some don't. An ECS salesman told me that it was better to use them with Athlon 1800 XP processors, and not much faster. He may have been talking about earlier revisions. The revisions currently at Fry's are version 5.
Comment removed based on user account deletion
Comment removed based on user account deletion
Comment removed based on user account deletion
On expensive Sun machines, people use *software* RAID. Software RAID is just as reliable, and it is easier to maintain, monitor, tune, and diagnose problems with.
Do the same on Linux. Just use the md driver. Its raid 5 is faster than any hardware raid 5 card you can buy.
Just put separate IDE drives on separate controllers. You're an idiot if you have any slave IDE drives. The performance hit is insane. RAID-5 them with md and you're set.
I'm not fat, and I'm not even big boned.
If tits were wings it'd be flying around.
I personally use Compaq Smart Array Controllers under NT, 2000, and 2003. They work great - never had a glitch with one. One problem is that they are *rather* long boards - might not fit in a case if it isn't full size. Compaq/HP has Linux stuff, I think, so they most probably have drivers to go along with it. Compaq's server drivers, on Windows, at least, have always been really high quality. I have no comment about their desktop & laptop drivers .
--Brian Desmond
4 or 8-port 3Ware controllers which have good Linux driver support. For RAID-1 they're as good as software RAID, for RAID-5 you should still use the 3ware controllers to give you hot-plug functionality but use Linux software RAID because it beats the hardware performance by a mile.
Matthew @ Bytemark Hosting