Slashdot Mirror


User: InsaneGeek

InsaneGeek's activity in the archive.

Stories
0
Comments
716
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 716

  1. Re:Is this filesystem immune to the "rhnsd factor" on Tux2: The Filesystem That Would Be King · · Score: 1

    Why don't you use a swap file on your Linux box

    dd if=/dev/zero of=/swap1 bs=1024 count=51200
    mkswap /swap1
    swapon -a /swap1

    On others (Solaris, Irix) it's even easier

    mkfile 50m /swap1
    swap -a /swap1

    And you've added an additional 50meg to swap.

    Your Win2k doesn't have a swap parition, it has a swap file, which is exactly what you are doing here. Pretty much every recent flavor of unix can create swap files and add them on the fly; some even have the very nice feature of virtual swap, so you never actually allocate physical disk (which is great for all those apps that reserve it but never use the 2gig of swap they want)

  2. Re:Strange on Inside the CueCat Hardware · · Score: 3

    Glad to see that you have deep pockets for court cases. No matter how silly a case is, unless you pay for competent lawyers you are screwed. Also once you start to put in motion actions that would allow you to counter-sue for legal fee's, it becomes a double-edged sword because then they can take their multi-million dollar lawyers and include that in their suit (where they couldn't before)

    You must be fairly young to believe that you can take on a company in a legal battle, without VERY competent representation. Even if you win, you are going to end up paying out lots of cash out of your very pocket. A company with millions of dollars can tie up a legal case for months... years... and you still have to pay your lawyer during that time.

    Things like this should NEVER be taken lightly, it may be the stupidest thing in the world; but if it gets to court 5-10k for a fees, lost work, etc. is nothing to them but that can break a lot of people I know. The best thing to do is to contact your well paid lawyer, under their guidance they will tell you to respond or not to. If they tell you a response it necessary, they draw up everything and you just put your signature on it. More than likely they will tell you to do nothing, and laughing at them, is the quickest way to get yourself into a lawsuit.

    Spelling & grammar checker off because I don't care

  3. Re:Where is the proof ? on Solution To DoS Attacks · · Score: 1

    How do we know that you aren't a fabrication? You could be some kind of bot. We have no proof that you actually are a human typing your statements. :)

  4. Re:Resizable Ramdisk + Unix = trouble on Other Uses For The Linux RAM Disk? · · Score: 1

    Hmm... well over the years of doing admin work, all the systems (at least Solaris, not SunOS) /tmp wasn't physical disk. Having had quite a number of friends who have worked in government organization, DOD, etc. companies, honestly it would not suprise me that it might be a hold over from SunOS (that's not supposed to be offensive, just an observation from what I've seen).

    If you look in the Solaris 2.6 System Admin. Guide, under the section about "Swap Space and the TMPFS File System", Solaris 2.x uses the swap for the temp filesystem. That doesn't say you can't change it, but for most of the apps we run on Sun (Oracle, etc.), having /tmp as a "virtual ram disk" gives us an obvious order of magnitude performance increase over spinning disk. Your applications may not get anything from /tmp & memory being together, so one can't make any blanket statements about setup for all configurations.

    Like I said I've never tried running quotas on the /tmp filesystem (I've beat my users into submission), and yes you'd get a bigtime slowdown, but off hand I can't think of another way to control the /tmp filesystem without a third-party reaper app. But you never know, quotas may not be that bad since you don't have to wait for the physical drive to give you information, one already has it in physical memory... I guess I know what I'll be trying on my new 420 tomorrow.

  5. Re:Resizable Ramdisk + Unix = trouble on Other Uses For The Linux RAM Disk? · · Score: 1

    The reason this was probably done is to improve performance of their MTA (which was probably failing because it had 500 other mail processes backed up waiting on I/O). For example each sendmail queue run it writes multiple tmp files, which just completely wails on the I/O (you're going to run out of I/O on a mail server before anything else). Stripping your queues over a cached raid system, will greatly improve performance, without having to resort to dangerous things like this. If you really NEED that type of performance using a SSD (a really expensive external ram disk, battery & disk backed up, that connects either using SCSI or F.C.) may be what you need since that will not eat up all system memory, etc., but it is EXPENSIVE with a capital E. Of course you can allways add swapfiles so that the system will stay up, albeit it will run extremely slow (imagine 6+ gig swapped out.. ouch), but it will stay up.

    I wouldn't be so quick to blame the SA, his company may not be willing to get him the hardware he needs to do his job properly, of course I don't know the situation and he may be a complete bumbling idiot, but you never know.

    Spelling and grammar checker off because I don't care

  6. Re:Resizable Ramdisk + Unix = trouble on Other Uses For The Linux RAM Disk? · · Score: 1

    Normally on a Solaris system /tmp isn't a filesystem at all, it is all available free memory (including physical & swap) at boot time. So if you fill up /tmp, you have filled up all your physical AND all of your swappable space; no more memory anywhere is available... crash ensues.

    You can go around this by changing your vfstab entries for swap, etc. but removing this is not suggested. There are lots of "Solaris'isms" that rely on quick access to these files, else your machine will go down to a crawl. I'm sure there also are quite a few 3rd party apps who rely on fast access to the /tmp/ filesystem, that if it were to be slowed down to spinning disk pain would ensue.

    You could probably put quotas onto the partition to prevent individual users from being bad players and taking a machine out. I've not had to do this, probably because I've always been very agressive in beating users into submission that, there are 2 temp areas on a system /tmp and /var/tmp and there better not be anything in /tmp that shouldn't go into /var/tmp.

    Spelling and grammar checker off because I don't care

  7. Re:law on Protecting Your Company While Protecting Privacy? · · Score: 1

    Yes it should and does, it was added for emphasis to plainly show to everyone my point.

  8. Re:law on Protecting Your Company While Protecting Privacy? · · Score: 1

    I never said it was a good or bad law, or that you had to snoop on you employees; but the poster gave as advice to someone wanting to protect themselves from lawsuits to NOT try to protect themselves and change the law, that's the part I have issues with. I wasn't saying to break the law either, but if one of his employee's does by accident, it's still his problem and often a multi-million dollar problem.

    You second paragraph is a mish-mosh of ranting. Do you believe those laws were put in place to only protect certain minorities? Do you think there should NOT be a law that says it's wrong to have a boss pull out his d*ck in front of his married secretary and give the option of being fired? I think the extent some of these laws are applied to is a problem, but what the laws were meant to protect by their creators I agree with.

    Actually you can fire someone for being rude. There is no law that says I can't fire you for being rude, dressing inappropriately, etc. I can not fire you for the usual race, religion, sexual orientation, etc. but anything else really is fair game.

  9. Re:Totally wrong solution on Protecting Your Company While Protecting Privacy? · · Score: 1

    Oh please, that is a pretty idealistic thinking, just get your voice out there the law will change.

    If that were as easy as you seem to make it be, I'm pretty damn sure that the fortune 100 companies who have a LOT more pull in local, state, federal government than government would have had it done already; instead of paying out the HUNDREDS of millions of dollars they have out for sexual harrasement settlements. In the YEARS that it will take to get any change with your "get you voice out there", that if you don't comply with the law you are in for a good old beating. I'm pretty sure that enforcement of that law is not going to listen to the excuse of "It doesn't apply to me, I'm getting my voice out."

    What you are saying is some of the worst advice I've ever heard, sure it will eventually work after years and years of fighting; but hey my company has to close shop because I didn't think that I should comply with the law that is currently on the books.

    I love the "ounce of prevention" statement, you first state that he shouldn't try to comply with the law to PREVENT a problem from happening, and then state "An ounce of prevention"... now that's comedy.

    Spelling & grammar checker off because I don't care

  10. Re:The proof is in the power... on Lord Of The Rings Being Rendered Under Linux · · Score: 5

    The thing is that people haven't touted Irix as being superior in performance for rendering, for doing the actual creation a SGI mips workstation will blow most anything away. Rendering is just pure and simple CPU, nothing really more than that, it doesn't take a big graphics pipe, just a big fast CPU, as long as the OS can get information from the drives to the CPU fast cool. Infact it running Linux has very little to it's performace, the x86 CPU is the one that is key there. Irix never was the slow point for rendering, but Irix only runs on mips and mips doesn't render fast. The operating system is MUCH less important than the CPU for rendering graphics (most of the time, never make a blanket statement).

    Many shops are using commodity boxes for rendering (running Linux, BSD, Solaris x86, or even NT), but for most of those same shops you'll have to pry their SGI workstations out of their cold dead hands.

    Linux and it's cost effective is a pretty useless point for most shops these days, the rendering software often costs 20+ times more than Irix does (lastest ver of Irix is $600), the savings comes in that you don't need that extra visual performance for rendering that the SGI system gives you, so you can use MUCH less expensive commodity hardware for that.

    Spelling and grammar checker off because I don't care

  11. Re:...only Twice? on Paying Twice For Windows · · Score: 1

    Ummm, just an observance...

    How many upgrades has linux gone through since 95? Linux is on an even bigger upgrade path, I call it version/patch of the day. I don't think a single week has gone by without some patch/upgrade coming out. Of course that's quite a bit cheaper upgrade, then from NT3-W2k

    Downtime, I'll mumble an agreement with that statement, and I'd even be more willing to state that out loud, once a journaled file system gets put into a standard distribution and is "officially" blessed.

  12. Re:What do they expect? on Napster Shut Down Until Trial · · Score: 1

    Actually my college being more liberal is something I do brag about, I feel that I am a much more well rounded individual, than some others narrow minded people that came from my hometown. During my time I received 2 degrees, and was working on a minor in theatre. I learned probably more than most people who go to prestigious universities because of the size of my college. For my CS degree program they combined 4 colleges to get 12 CS professors (as long as you don't mind a 30 min commute) so I had a large knowledge base compared to the class size (largest was around 15 people, smallest was 2). I feel that I learned more at this small college than I could at a large university. One professor specialized in compiler construction, another in algorthmic research, another in software engineering, another in.... I had a damn good time.

    Ummm... you completely side-stepped what I asked, again re-read my post and tell me where I'm false. That's all I'm asking, tell me where I'm incorrect. And for the facts, I didn't see a single one relevent one in your paragraph. All you did was say RvW and C.P. turned out that way because of public opinion, hell using your way of providing facts I could say Jeffry Dahmer was only convicted because public opinion was against him. And if I use your method of "truth" it's because that's the way it is so it's a fact.

    I'm not arguing that RvW wasn't influenced by public opinion, but for some damn reason you are trying to put words in my mouth, that I said public opinion is supposed to influence the courts. Then you are trying to say that I'm whining and moaning and bitching about SOMETHING I NEVER SAID. Wanna continue trying to spread lies about what I said, or if you are up to it why don't you in your next post actually

    1) Respond to my second request to where I said courts never get influenced by public opinion?

    2) Find a single news story that said the RvW judge wasn't fully on the up and up (I'm pretty damn sure the anti-abortionists wouldn't have let that stand).

    For future reference a fact is something that can be proven NOT a single statement by someone.

    Here's a fact for you, remember that dogs are animals, and they watch tv, and listen to the radio, and bark at other dogs, and they listen to what people say. You can whine and moan and bitch all you want about how it shouldn't happen, but those dogs are just studying us to take over the world. That about right on how you prove your facts???

  13. Re:What do they expect? on Napster Shut Down Until Trial · · Score: 1

    Exactly where does Roe v Wade fit in here???? Where exactly did the court ignore law and go with public opinion? Wanna step up and back your own statements? Did the courts not enforce the laws properly? The court system correctly chose the law that was of the higher order, public opinion didn't magically pop out a new law, they used an existing law on the books to over-rule another law, no public opinion if the law didn't already exist, I'm pretty damn sure that R v W would have gone the other way.

    Where EXACTLY did I state that public opinion can not sway a trial? Please point it out to me, I believe I said that the judicial system is MEANT to. I never said it never happens, nobody would believe that opinion can never sway a court; but I never DID SAY THAT IT WOULD NOT. Again I will state the FACT that the judicial is not SUPPOSED to listen to public opinion; it is to interpret the laws that are already made and base their judgements solely on that.

    Oh, and I went to college, (a liberal arts college in fact, good times...) and I new quite a long time before then that the world never revolves along the idealistic goals that it is supposed to abide by. On a side note, I bet my college was probably even more liberally eye opening then yours, mine actually paid to send students to Canada, to escape the draft during Vietnam. (top that hehehe.)

    If you want to call my response an attack then fine whatever. If someone is going to respond sarcastically to me, with absolutely no true facts or is so completely wrong in such a simplistic fasion... then yes, I tend to respond in a curt way.

    Spelling and grammar checker off because I don't care

  14. Re:What do they expect? on Napster Shut Down Until Trial · · Score: 1

    Hmm... Silly me, I thought that U.S. laws were made by the legislature who are elected by the people and are responsible to those people.

    The judicial system conversly is meant ONLY to enforce those laws, it is NOT there to make laws or use their power to reflect the views of the people. The courts cannot use public opinion in their views, they are to only look at the laws written by the branch of government elected by the people and execute those laws blindly (good or bad).

    U.S. laws are meant to reflect the views of the people, but that does not mean that the court system is to go with public opinion. So in a pure sense what your sarcastic statement is right; but when you included my quote, it showed that you need to go back a few grades in high school because you aren't smart enough to be out yet.

    Grammar & spell checker off because I don't care

  15. Re:What do they expect? on Napster Shut Down Until Trial · · Score: 1

    Is this as silly as say, when Microsoft had an injuction put against their Java implementation months before the trial ended? I'm not for Microsoft, but you gotta take the good with the bad, the court system isn't supposed to be swayed by public opinion.

    They presented enough evidence that a judge found the case reasonable to stop Napster while the trial is in progress. Truely I want the government to work this way, for an example if I was wronged; let's say someone dumped my GPL'd code into a closed commercial product. If I was able to present enough evidence to the jude that my case was resonable and that the continued use of that was detrimental to me, I'd like to see that use stopped. Cases may run for years and years, if someone is able to reasonably show that another one is in the wrong; I'd prefer a temporary stop be put in place.

    But hey that's just me, I apply my statements to cases in general not to just napster; like I said take the good with the bad.

    spelling & grammar checker off because I don't care

  16. Re:Raid 5 for a database? You must be kidding. on 30+ GB Databases On Unix? · · Score: 1

    Actually you want a combination of raid 0 and 1. Raid 0 will get you the performance and 1 will get you the redundancy you want. Raid 5 is VERY bad for any types of writes, the previous poster stated it very well, for reads it's OK, for writes it completely sucks.

    I have yet to see a large databaes running Raid 5, small ones yes, ones that never change yes, but anything that gets updated regularly or is big enough to actually require something like Oracle/Sybase/Informix Raid 5 doesn't cut it.

    Another thing a NetApp uses Raid 4 for their WAFL filesystem, snapshot is kinda cool as long as you aren't doing lots of writes (you can completely overwrite all your snapshot space, and then you're SOL). I still have issues with running a database over NFS even with gig ethernet directly attached. Of course, I still have problems with running most anything production over NFS (burned in the trenches geezer).

  17. As long as the ISP feels it's necessary on What Kind Of Logs Should ISPs Keep? · · Score: 1

    Having been in the ISP trade for a number of years until I finally escaped awhile back; the best advice that I have for ISP's is to keep it around as long as the ISP thinks it's necessary. Let me add a note to that statement, when you are determining "necessary" be prepared to have a government agency outside your door someday and take weeks to months sifting through the data. We didn't keep any network level information (we could care less about routing), but we did record all dial-in access authentication for 6 months.

    We recorded: login/logoff times, ipaddresses, bytes in/out, and on digital lines the CID

    We didn't watch what you did on our network, but we did record the initial logins for general accountability of our customers, tracking of which pop sites need modem/bandwidth upgrades (lots of other tools used here, but the more data the easier to get financing for upgrades). Our necessary was 6 months, for others it may be less, (we were keeping somewhere around a 1/4 to 1/2 million dialin records a day).

    We had a very formalized schedule and I suggest any ISP to get some form of it onto paper. We talked with many different government agencies and the resounding word was that the ISP business is unregulated, they don't care how long the records are kept around; but if they were to contact us and someone tells them 2 weeks, another tells them 6 months, and another tells them they never get deleted, that they are going to show up on your doorstep to do an inperson hello. As long as everybody gives them the same answer then they are fine with you giving them the logs, if they smell a hint of something fishy (even if it's only a simple communication error) they are coming in with a warrant to seize your equipment, and will go through all your equipment at their leisure where you may or may not allow your business to run.

    Spelling & grammar checker off because I don't care

  18. Re:Then they can't use NT or W2k on Are Linux Transactions Slower Than Win2k's? · · Score: 1

    You shoulda looked at my "wait for a full year", which I know more than a couple of companies do for service packs (hot fixes are a different story). I don't work on the NT side of the world in our organization, but I do know that we don't have any 2k where I work.

    And your "Service Pack" statement illustrates my point even better, Microsoft RUSHED the pack out, instead of taking the time to certify the product around. Businesses don't rush out and upgrade very often, at least ones that stay in business don't.

  19. Re:6.1? on Are Linux Transactions Slower Than Win2k's? · · Score: 1

    Yes, but businesses aren't into the upgrade every week (or even every 4 months) paradyne. Some wait a full year for an OS release to mature before they roll it out. One can put along with this, that many 3rd party vendors have product support listed only for Redhat 6.1 (it should work with 6.2 but if it breaks your on your own). That is the biggest problem with the fast pace of Linux upgrades, vendors don't have the luxury of 20 billion bug breakers for their code, they have to spend lots of time verifying that their code works against any upgrades. An example of this is our EMC EDM backup system is only "certified" to work with 6.1, we pay lots in support yearly (begins with mill..), so we don't mess around with not supported options.

    So from a business server perspective 6.1 is good, for a home or workstation using 6.2 much better.

  20. Re:Indigo 2? on NetBSD Ported To SGI 02 · · Score: 1

    I'd actually have to disagree with you on the insecure Irix, in the past yes; but not in the past couple of years. If you checkout securityfocus.com and look at the vulnerability stats, Irix & NetBSD are at about the same level (Irix had fewer then NetBSD, Linux, Solaris, but more the OpenBSD last year).

    I'd say that if they can get X going on your IndigoII then go with NetBSD, Irix is great for big SMP & Numa boxes but having all of that additional support can make it drag for single CPU boxes.

  21. Re:Try $500 for an Indy on NetBSD Ported To SGI 02 · · Score: 1

    I purchased my IndigoII about 3 years ago with 20" monitor for $1800, Indy's are currently going for $500, and my Indigo is ~$900 now.

    For prices, checkout

    www.reputable.com
    www.recurrent.com
    www.mce.com

    I'm still waiting to get a "good" port of SGI Hardhat Linux for it, but BSD is great in my eyes I'll see if I can spin it up on one of our 10k Origin boxes though....

  22. Re:Maybe we'd pay if... on NetSol To Do Domain Name Auctions · · Score: 1

    Why do you seem to think that this is NSI's problem, and not your companies? Your company CHOSE to use the least secure method instead of using the pgp or crypt-pw option. NSI didn't force you to use an unsecure method, they give you 3 options. With all of the press these cracks have gotten over the past few years (and everyone of them mentioning using the secure option would have prevented it), someone should have pulled their head out.

    I'd actually say that I'd think twice about doing business with your company. After being burned once, and then not requesting one of the usual secure methods... sorry, but that is completely idiotic.

    Spelling & grammar checker off because I don't care.

  23. Re:I won't buy one just yet. on Transmeta To Unveil New Notebooks Next Week · · Score: 2

    An even more interesting thought, is what about programs specifically attacking the processor code? Look at all of the bios vir[ii|uses|us's] and such that are out currently, and now think of ones specifically going after the cpu... imagine all of the nasties that could occur: slow the computer down by 4x or give an incorrect number every millionth computation. There could be lots of fun things done with the flexibility that Trans. is giving to the consumer for both good AND bad.

    No matter how hard they try to make that code unchangeable someone's going to get around it. For an example, look at how many different programs got around the Intel ID option that was supposedly impossible to change by a rogue program.

    Anyway that was my thought and mine alone

  24. Re:RAID != speed on Linux Now Supports Ultra ATA/100 · · Score: 1

    Anytime you add raid you are adding performance costs. A stripe (best performing raid) set of 8 drives will get slightly less performance then 8 individual drives (but you now have 8x space on one slice). True striping cost is so low that it really doesn't matter, but the cost of CPU time is there (yes hardware raid has CPU time too).

    What you are saying is so truely wrong, that I'm not sure if it's a good troll (power to you if so), or you haven't thought things through at all. What you are suggesting is completely pointless, raid gets you additional levels of protection, raid 0 less but you get to combine smaller drives to get a bigger drive that is faster than one single drive. That's why it's Redundant Array of Inexpensive Disks.... RAID

  25. Re:This looks inacurate... on SANS Releases Top Ten Exploits · · Score: 1

    Whoops, when I said freaks I mean that for the hardcore, Linux can do no wrong, type people. I've ran into more than a few people who seem to think of Linux as a religion, and that any comments about improvement, or other OS's doing something better is heresy.

    Spelling & grammar checker off because I don't care