I wish the US hadn't done such stupid things to get people so angry at us that they feel their only recourse is to blow up buildings.
And this justifies blowing buildings in what way?
war in a country that had nothing to do with any terrorist attacks
Regardless of the truth of this steatement, at least a terrible dictator that has killed millions of innocent people is now behind bars. I think it's worth it, and I bet most of the troops over there think it's worth it.
Quite true, if it is super critical, a followup call is a must. Nevertheless, a hard drive should not lose data that was reported by the kernel to be synced to disk.
Yeah some more expensive controllers can do that, but I don't think you'd hook that up to a system powered by a 500Mhz CPU.:)
One other problem with write cache (that maybe the powered RAM cards fix) is when rebooting - I have seen a system go down for reboot, and hits the POST initialization before the controller is finished syncing the disks... and then it gets reset. Poof! lost data.:(
The advice I was trying to give was to not discount Software RAID on account of RAM or CPU - it sounds like plenty for that arrangement. It also sounds like they are not building the very best system, so most if this wont really have any relevance.
I hadn't considered the fact data smaller than the stripe size, but before you pull out your flaming gins look at the parents of my posts to see how far off they were... Software RAID is perfect for the OP's problem, no need to use hardware RAID and Linux is most certainly up to the task.
Since you seem to know so much... isn't it likely, though, for many operations, the data is already in cache? For update operations anyway, the data was probably read recently, which would have had the entire block. In this case, extra reads may not be necessary.
Ah ok, I see what you are saying... although, chances are that the whole sector has already been read previously and is already in cache anyway.
For example, when the data was read prior to being modified.
The only time this would be a problem is when a whole new file is being written. Even then, the sector may be on the end of a free space inode chain, and so the kernel know no other data is on that sector, and it can do what it wants. Depends on the filesystem, I guess.
Ok, network could be a bottleneck, if the disk IO gets high enough. I think, though, if this is a 500Mhz system, it is likely to need more attention on the PCI/ATA bus than on the network.
Both network and IO are to be worried about before CPU.:)
It's a poor solution for raid, since if the OS goes, there goes your raid. If you use hardware, at least it'll autodetect.
Um, that's bogus. If your OS goes (probably due to hardware?) then you can simply put the drive in a new computer (same basic master/slave setup) and away it goes. Linux knows how to detect its own RAID arrays!
OTOH, if you have a hardware RAID, good luck getting tech support, especially if they no longer carry that board, or have gone out of business altogether.
At least with Software RAID, your data is not stuck in a proprietary format.
Um, no. Since we are writing, we already know what the data is. Just write. No reads.
RAM is helpful on ly in the sense that it can cache data and make things appear faster if the data is already available in cache. It won't really help it read/write data any faster.
If a drive goes out, write performance is inchanged, as the XOR operation must be done no matter what.
Read performance depends - it's just an XOR operation, which is not very difficult. A 500Mhz CPU will still be mostly idle even in degraded mode. Some implementations *could* do the XOR in regular mode too, to check for data errors, in which case, no performance is lost in degraded mode anyway.
the card can tell the OS everything is done being written and then flush it out of cache at its convinience.
Which is absolutely horrible. This violates protocol - mail MTA's demand that data is written to disk before they acknowlege delivery. They get this from the confirmation from the kernel, but if the disk array lies about it, a power failure could lose data even though the kernel assumed it had bee synced properly.
IDE interfaces, and some SATA interfaces, are not designed to be hotplugged. There are special electrical circuits on hot pluggable scsi drives.
Once you yank the plug on an IDE, you *must* power down the system before plugging it back in. Sometimes yanking the plug hangs the kernel.
Power down the system, pull a drive, and then start it up. It should detect the missing drive on bootup. Alternately, follow the instruction on the LINUX RAID HOWTO. You have read that, haven't you?
If all it does is serve files, it should do fine. The 500Mhz is not going to be a factor at all, in fact, the CPU will be idle most of the time. The real thing to optimize in a file server is the ATA bus speed and hard drive latency.
RAID 5 hardware tends to be rather expensive, and most RAID hardware tends to be "pseudo hardware", the drivers for the raid card make the CPU do the actual work anyway. Your 500Mhz CPU is faster than all but the most expensive RAID controllers anyway.
Stick with Linux RAID. It knows how to do it better.
The idea is that in order to write data to any sector on one of the drives, the sectors from six of the other drives need to be read, all XOR'd together, and then the result written to the remaining drive.
Your logic eludes me. The blocks do not need to be read, as we are in the process of writing. We already have the data, because we are writing, so why would we re-read the data?
Furthermore, block sizes default to 4k, though you could go to 8k or 32k block size. At any rate, you don't need a gig of RAM to handle this.
Finally, XOR is not that expensive of an operation, and a 500Mhz CPU is going to be able to handle that faster that any but the most expensive controller cards.
So unless you are actually a RAID kernel developer, I don't buy your story.
A virus that destroys a computer quickly is also a very poor vector, because it doesn't live long enough to infect others. The most explosive viruses have been ones that did no other damage than to reproduce. Of course, this causes DoS's on networks, but the computer is still happily infected.
headquarters refused to allow me to connect my laptop to their network unless I could demonstrate that a reputable virus scanner was checking my machine at least daily.
ClamAV gets updated faster than the major AV companies, and some really neat matching algorithms match mutations before specific signatures are released. Very reputable.
This is exactly the direction we are planning to go with Maia Mailguard, plus features such as tarpitting, network reporting, and p2p associations. It's going to take a while to get there, though.
Go with Postix/amavisd-new/clamd/spamassassin/Maia Mailgaurd. Can integrate with sql or ldap for account maintenance. Can scale across multiple computers. Very slick.
Sorry, I've tried PageMill, and FrontPage, and Netscape and Mozilla built-in editors, and even MS Office's HTML editing. Don't like them. They all generate bulky, messy code, hard to tweak, impossible to really control.
Have you tried Quanta Plus? It lets you hand type all that stuff, but also keeps track of tags and doctypes for you. Pretty cool, except for the fact that it crashes immediately on startup for me right now.:/
And if you think hand-coded HTML is unpretty, somehow, visit http://www.worship-live.com for what you can do without an editor.
eek. I would have some issues with your style of presentation - there's too much stuff on one page. Nevertheless, you are right, in that your hand coding is clean. Your code is nearly xhtml compliant, however, and you do make use of CSS, so it would not take much to convert to xhtml. I'm not clear if that was you argument anyway.
As I said there's too much stuff on one page - I'd break it up some. I kept looking, however, because I'm actually looking for this type of software for our church. Amazing where you find leads sometimes, isn't it?!
Slashdot Mirror
And this justifies blowing buildings in what way?
war in a country that had nothing to do with any terrorist attacks
Regardless of the truth of this steatement, at least a terrible dictator that has killed millions of innocent people is now behind bars. I think it's worth it, and I bet most of the troops over there think it's worth it.
so I'm not the only one who sees the potential for a tie... I didn't even realize that it was possible before. I got a funny feeling about this...
I think I heard that the decision could then go to the House of Representatives for vote. I'm not sure about that, though.
Of course, if you have ad-block installed... :P
Quite true, if it is super critical, a followup call is a must. Nevertheless, a hard drive should not lose data that was reported by the kernel to be synced to disk.
Yeah some more expensive controllers can do that, but I don't think you'd hook that up to a system powered by a 500Mhz CPU. :)
:(
One other problem with write cache (that maybe the powered RAM cards fix) is when rebooting - I have seen a system go down for reboot, and hits the POST initialization before the controller is finished syncing the disks... and then it gets reset. Poof! lost data.
The advice I was trying to give was to not discount Software RAID on account of RAM or CPU - it sounds like plenty for that arrangement. It also sounds like they are not building the very best system, so most if this wont really have any relevance.
I hadn't considered the fact data smaller than the stripe size, but before you pull out your flaming gins look at the parents of my posts to see how far off they were... Software RAID is perfect for the OP's problem, no need to use hardware RAID and Linux is most certainly up to the task.
Since you seem to know so much... isn't it likely, though, for many operations, the data is already in cache? For update operations anyway, the data was probably read recently, which would have had the entire block. In this case, extra reads may not be necessary.
Ah ok, I see what you are saying... although, chances are that the whole sector has already been read previously and is already in cache anyway.
For example, when the data was read prior to being modified.
The only time this would be a problem is when a whole new file is being written. Even then, the sector may be on the end of a free space inode chain, and so the kernel know no other data is on that sector, and it can do what it wants. Depends on the filesystem, I guess.
Ok, network could be a bottleneck, if the disk IO gets high enough. I think, though, if this is a 500Mhz system, it is likely to need more attention on the PCI/ATA bus than on the network.
:)
Both network and IO are to be worried about before CPU.
It's a poor solution for raid, since if the OS goes, there goes your raid. If you use hardware, at least it'll autodetect.
Um, that's bogus. If your OS goes (probably due to hardware?) then you can simply put the drive in a new computer (same basic master/slave setup) and away it goes. Linux knows how to detect its own RAID arrays!
OTOH, if you have a hardware RAID, good luck getting tech support, especially if they no longer carry that board, or have gone out of business altogether.
At least with Software RAID, your data is not stuck in a proprietary format.
Um, no. Since we are writing, we already know what the data is. Just write. No reads.
RAM is helpful on ly in the sense that it can cache data and make things appear faster if the data is already available in cache. It won't really help it read/write data any faster.
If a drive goes out, write performance is inchanged, as the XOR operation must be done no matter what.
Read performance depends - it's just an XOR operation, which is not very difficult. A 500Mhz CPU will still be mostly idle even in degraded mode. Some implementations *could* do the XOR in regular mode too, to check for data errors, in which case, no performance is lost in degraded mode anyway.
the card can tell the OS everything is done being written and then flush it out of cache at its convinience.
Which is absolutely horrible. This violates protocol - mail MTA's demand that data is written to disk before they acknowlege delivery. They get this from the confirmation from the kernel, but if the disk array lies about it, a power failure could lose data even though the kernel assumed it had bee synced properly.
IDE interfaces, and some SATA interfaces, are not designed to be hotplugged. There are special electrical circuits on hot pluggable scsi drives.
Once you yank the plug on an IDE, you *must* power down the system before plugging it back in. Sometimes yanking the plug hangs the kernel.
Power down the system, pull a drive, and then start it up. It should detect the missing drive on bootup. Alternately, follow the instruction on the LINUX RAID HOWTO. You have read that, haven't you?
If all it does is serve files, it should do fine. The 500Mhz is not going to be a factor at all, in fact, the CPU will be idle most of the time. The real thing to optimize in a file server is the ATA bus speed and hard drive latency.
RAID 5 hardware tends to be rather expensive, and most RAID hardware tends to be "pseudo hardware", the drivers for the raid card make the CPU do the actual work anyway. Your 500Mhz CPU is faster than all but the most expensive RAID controllers anyway.
Stick with Linux RAID. It knows how to do it better.
The idea is that in order to write data to any sector on one of the drives, the sectors from six of the other drives need to be read, all XOR'd together, and then the result written to the remaining drive.
Your logic eludes me. The blocks do not need to be read, as we are in the process of writing. We already have the data, because we are writing, so why would we re-read the data?
Furthermore, block sizes default to 4k, though you could go to 8k or 32k block size. At any rate, you don't need a gig of RAM to handle this.
Finally, XOR is not that expensive of an operation, and a 500Mhz CPU is going to be able to handle that faster that any but the most expensive controller cards.
So unless you are actually a RAID kernel developer, I don't buy your story.
A virus that destroys a computer quickly is also a very poor vector, because it doesn't live long enough to infect others. The most explosive viruses have been ones that did no other damage than to reproduce. Of course, this causes DoS's on networks, but the computer is still happily infected.
In my experience, it should be at the top of the list.
I opened up pop3://user@host.tld in konqueror and got a directory listing. I can't write any files though.
imap://... gets a directory listing, but I can't see any messages, and can't write. This could be really cool if finished.
ClamAV
ClamAV gets updated faster than the major AV companies, and some really neat matching algorithms match mutations before specific signatures are released. Very reputable.
This is exactly the direction we are planning to go with Maia Mailguard, plus features such as tarpitting, network reporting, and p2p associations. It's going to take a while to get there, though.
Learn either LILO or GRUB like your life depends on it.
And for the love of Pete, If you compile a new kernel, DONT OVERWRITE THE OLD KERNEL!!!!!
Set up a new menu entry, so you can always failsafe back into the old kernel. I don't know how many times I've seen this done.
Indeed, Here's the logfile entries that showed up:
/home/mortonda/public_html/htdocs/ok.txt
66.35.250.150 - - [01/Sep/2004:14:15:29 -0500] "GET http://yro.slashdot.org/ok.txt HTTP/1.0" 404 1034 "-" "libwww-perl/5.76"
[Wed Sep 01 14:15:29 2004] [error] [client 66.35.250.150] File does not exist:
Sep 1 14:15:17 freedom kernel: SFW2-INext-DROP-DEFLT IN=eth0 OUT= MAC=00:50:fc:e3:9a:cc:00:04:5a:f5:c3:44:08:00 SRC=66.35.250.150 DST=192.168.1.151 LEN=60 TOS=0x00 PREC=0x00 TTL=51 ID=17492 DF PROTO=TCP SPT=33207 DPT=1080 WINDOW=5840 RES=0x00 SYN URGP=0 OPT (020405B40402080A1D1CAC6D0000000001030300)
Sep 1 14:15:20 freedom kernel: SFW2-INext-DROP-DEFLT IN=eth0 OUT= MAC=00:50:fc:e3:9a:cc:00:04:5a:f5:c3:44:08:00 SRC=66.35.250.150 DST=192.168.1.151 LEN=60 TOS=0x00 PREC=0x00 TTL=51 ID=17493 DF PROTO=TCP SPT=33207 DPT=1080 WINDOW=5840 RES=0x00 SYN URGP=0 OPT (020405B40402080A1D1CAD990000000001030300)
Sep 1 14:15:20 freedom kernel: SFW2-INext-DROP-DEFLT IN=eth0 OUT= MAC=00:50:fc:e3:9a:cc:00:04:5a:f5:c3:44:08:00 SRC=66.35.250.150 DST=192.168.1.151 LEN=60 TOS=0x00 PREC=0x00 TTL=51 ID=29239 DF PROTO=TCP SPT=33208 DPT=3128 WINDOW=5840 RES=0x00 SYN URGP=0 OPT (020405B40402080A1D1CAD990000000001030300)
Sep 1 14:15:23 freedom kernel: SFW2-INext-DROP-DEFLT IN=eth0 OUT= MAC=00:50:fc:e3:9a:cc:00:04:5a:f5:c3:44:08:00 SRC=66.35.250.150 DST=192.168.1.151 LEN=60 TOS=0x00 PREC=0x00 TTL=51 ID=29240 DF PROTO=TCP SPT=33208 DPT=3128 WINDOW=5840 RES=0x00 SYN URGP=0 OPT (020405B40402080A1D1CAEC50000000001030300)
Sep 1 14:15:23 freedom kernel: SFW2-INext-DROP-DEFLT IN=eth0 OUT= MAC=00:50:fc:e3:9a:cc:00:04:5a:f5:c3:44:08:00 SRC=66.35.250.150 DST=192.168.1.151 LEN=60 TOS=0x00 PREC=0x00 TTL=51 ID=61874 DF PROTO=TCP SPT=33209 DPT=8000 WINDOW=5840 RES=0x00 SYN URGP=0 OPT (020405B40402080A1D1CAEC50000000001030300)
Sep 1 14:15:26 freedom kernel: SFW2-INext-DROP-DEFLT IN=eth0 OUT= MAC=00:50:fc:e3:9a:cc:00:04:5a:f5:c3:44:08:00 SRC=66.35.250.150 DST=192.168.1.151 LEN=60 TOS=0x00 PREC=0x00 TTL=51 ID=61875 DF PROTO=TCP SPT=33209 DPT=8000 WINDOW=5840 RES=0x00 SYN URGP=0 OPT (020405B40402080A1D1CAFF10000000001030300)
Sep 1 14:15:26 freedom kernel: SFW2-INext-DROP-DEFLT IN=eth0 OUT= MAC=00:50:fc:e3:9a:cc:00:04:5a:f5:c3:44:08:00 SRC=66.35.250.150 DST=192.168.1.151 LEN=60 TOS=0x00 PREC=0x00 TTL=51 ID=49346 DF PROTO=TCP SPT=33211 DPT=8080 WINDOW=5840 RES=0x00 SYN URGP=0 OPT (020405B40402080A1D1CAFF10000000001030300)
Sep 1 14:15:29 freedom kernel: SFW2-INext-DROP-DEFLT IN=eth0 OUT= MAC=00:50:fc:e3:9a:cc:00:04:5a:f5:c3:44:08:00 SRC=66.35.250.150 DST=192.168.1.151 LEN=60 TOS=0x00 PREC=0x00 TTL=51 ID=49347 DF PROTO=TCP SPT=33211 DPT=8080 WINDOW=5840 RES=0x00 SYN URGP=0 OPT (020405B40402080A1D1CB11D0000000001030300)
Ok, I'm curious...
Go with Postix/amavisd-new/clamd/spamassassin/Maia Mailgaurd. Can integrate with sql or ldap for account maintenance. Can scale across multiple computers. Very slick.
Have you tried Quanta Plus? It lets you hand type all that stuff, but also keeps track of tags and doctypes for you. Pretty cool, except for the fact that it crashes immediately on startup for me right now.
And if you think hand-coded HTML is unpretty, somehow, visit http://www.worship-live.com for what you can do without an editor.
eek. I would have some issues with your style of presentation - there's too much stuff on one page. Nevertheless, you are right, in that your hand coding is clean. Your code is nearly xhtml compliant, however, and you do make use of CSS, so it would not take much to convert to xhtml. I'm not clear if that was you argument anyway.
As I said there's too much stuff on one page - I'd break it up some. I kept looking, however, because I'm actually looking for this type of software for our church. Amazing where you find leads sometimes, isn't it?!