Mostly OT: Tekram caching IDE controller
Anyone used the DC690 under Linux? Its a PCI card, 2 IDE busses, no CDROM support, with 4 30pin SIMM slots. Supposedly does mirroring, all of which is transparent to the OS. It doesn't require any DOS drivers. Any pointers would be apprectiated. -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
Re: FAQ update
Luca Berra wrote: > i'd add: dont use netscape to fetch patches from mingo's site, it hurts > use lynx/wget/curl/lftp Works fine for me. -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
Re: FAQ - a suggestion
How about just putting something in like: "Uncompressing the patch is beyond the scope of this document." -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
Re: FAQ
Marc Mutz wrote: > My tar cannot use bz2-compressed unless used with > --use-compress-program=bzip2. so that line sould probably read "bzcat > kernel-2.2.16.tar.bz2 | tar xf -". Also the only tar I saw that knows > bzip2 is slackware's and it uses the '-y' switch for that. I never saw > the '-I' switch for tar and my 'info tar' does not list it. Bottomline: > Your tar is too customized to be in a FAQ. How about both options? The tar that comes with RH6.2 does this just fine... Ed -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
OT: best cross-OS filesystem
Sorry to waste bandwidth, but I'm looking at a way for better cross-OS performance on my "shared" partition - are there ext2fs drivers for NT somewhere, or maybe hpfs drivers for NT? I have some very large directories with 100's of files, and I want to be able to get in and around them easily...
Re: Failure autodetecting raid0 partitions
Wow, an email CCed to Linus himself! *faint*
Re: speed and scaling
Boy I don't know who was screaming about PCI bandwidth, but : 1) I was the first person to mention it, weeks ago, so its no news flash, and you're no genius for thinking of it. 2) Stop screaming. 3) Be informed. P.S. Let us know what happens, Seth, this sounds like a cool project! Ed
Abit KA7 + RAID
I saw a blurb somewhere about this board offering built in RAID 0 and 1, a BIOS thing. Is this just more WinRAID, like the Promise Fasttrak? -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
Re: speed and scaling
Seth Vidal wrote: [monster data set description snipped] > So were considering the following: > > Dual Processor P3 something. > ~1gb ram. > multiple 75gb ultra 160 drives - probably ibm's 10krpm drives > Adaptec's best 160 controller that is supported by linux. [snip] > So my questions are these: > Is 90MB/s a reasonable speed to be able to achieve in a raid0 array > across say 5-8 drives? While you might get this from your controller data bus, I'm skeptical of moving this much data consistently across the PCI bus. I think it has a maximum of 133 MB/sec bandwidth (33 MHz * 32 bits width). Especially if (below) you have some network access going on, at near gigabit speeds.. you're just pushing lots of data. > What controllers/drives should I be looking at? See if there is some sort of system you can get with multiple PCI busses, bridged or whatnot. > And has anyone worked with gigabit connections to an array of this size > for nfs access? What sort of speeds can I optimally (figuring nfsv3 in > async mode from the 2.2 patches or 2.4 kernels) expect to achieve for > network access. I've found vanilla nfs performance to be crummy, but haven't played with it at all. Ed -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
Re: attempt to access beyond end of device
Rop Slijkerman wrote: > > Doe anyone know what this message means? > > Jun 13 23:11:58 hoegaarden kernel: attempt to access beyond end of > device > Jun 13 23:11:58 hoegaarden kernel: 08:21: rw=0, want=251956997, [SNIP] I don't think this is a raid error, I've gotten it on a single-disk system - typically where there is some problem with the drive partition table. In my case, I had let Win95 partition the drive, which caused some incompatabilities with the way NTWS and Linux 2.2 were looking at the partitions. This was a 20Gb Seagate. -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
Re: 2.2.16 RAID patch
Marc Haber wrote: > > On Mon, 12 Jun 2000 14:34:59 +0200 (CEST), Ingo Molnar <[EMAIL PROTECTED]> > wrote: > >the latest 2.2 (production) RAID code against 2.2.16-final can be found > >at: > > > >http://www.redhat.com/~mingo/raid-patches/raid-2.2.16-A0 > > > >let me know if you have any problems with it. > > A kernel patched this way doesn't build with Debian's kernel package. > Complains "The version number 2.2.16-RAID is not all lowercase. Stop." > > Could this be changed to 2.2.16-raid for future versions or should I > better get in touch with kernel-package's maintainer? Or just get a real kernel from kernel.org. None of RedHat's kernel-o-matics patch cleanly either, so if you ever need ACL support, ATA66 support, RAID support, simpler access to kernel patches, then just install the real thing from kernel.org. Seems a bit uncool to get the coders to create (unless its purely the case of the patch filename) multiple or distro specific stuff. -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
HW? RAID on the HPT370
HighPoint has a new chip, the HPT370, which will probably make it to an ABit motherboard near you fairly soon. It's an ATA100 IDE controller that, according to the web fluff - (quote) With the inclusion of Hot Swap capabilities, user will have the ability to remove IDE/ATAPI devices from the system without shutting the system down. Another important feature is RAID 0, 1 and 0+1 support, which allows users to connect multiple hard drives for data mirroring and striping applications for performance and data-redundancy concern. (unquote) So, what does this mean - "RAID support". Has anyone seen one of these beasts yet? Ed
Re: best ATA66 card for linux ??
Might want to wait until people ship ATA-100 controllers, like the HPT370. -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
Re: ATA66 Raid
Paul Jakma wrote: [SNIP] > if you think any company would build a card with a general purpose > CPU, (at least $15, if not ~$150 for a StrongARM), FlashRAM, DRAM, > disk controller (IDE or SCSI), I/O bus interface (eg PCI bridge) then > deliberately cripple it so that it could only act as a drive > controller then you are mad. :) Not if it sucked - the point is that it could be done fairly cheaply, with cheap components. Would you get high performance? No. Would it benefit from doing much of the processing on the host CPU? Sure. Would the drivers take advantage of it? Uh-huh. Would Promise save on inventory and fab costs by having essentially one product? Yes. I'm not saying that they DO, or WOULD, but I wouldn't be at all surprised. repeat after me: I'm a know-it-all who likes to flame. -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
Re: ATA66 Raid
"m. allan noah" wrote: > > i will point this out AGAIN. the promise fastTrak is NOT hardware raid. did > you really think 30 seconds with a soldering iron and a 7 cent resistor could > make a hardware raid device from a 20 dollar ide card? Well, yes, if Promise intentionally broke the card. Case in point - MS with their fabled "NT Workstation is really really different than NT Server" yet 3 registry entries make one into the other. YES, it's software RAID. But YES, it's not out of the realm of possibility that a 7 cent resistor could make the normal card into a real RAID card. -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
Re: swap on RAID - swapFILE?
"m. allan noah" wrote: > > no. just to recap the discussion that has been occurring on and off this list > as i understand it: > > 1. it is NOT safe to swap to a raid partition while reconstuction is occuring. >the general consensus is that this is true whether or not the swap >partition is the one syncing. therefor, most folks seem to avoid swapon >until reconstruction is done on all raid partitions. > > 2. it is not safe to swap to a swapfile on a raid partition while >reconstuction is occurring. linux does not swap to a file exactly. it >actually swaps to the blocks occupied by that file. this is potentially >plagued with the same problems a case #1. Hmm, I assumed that swapping to a file would have to go through the filesystem, as opposed to a standard raw device. -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
swap on RAID - swapFILE?
What if you use a swapfile for swap? And place that FILE on the RAID array? -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
Re: Problem with RAID5 - corrupt files
What kinds of disks are you using? I'm seeing occasional corruption on my Seagate 20MB ATA66 barracuda, and have not ruled out the possibility of unstable ATA66 mode. -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
RAID vs. tiobench vs. 2xIDE disks
I have 2 IDE disks, each as master on their own channel. One is significantly faster than the other, benchmarking scores of 15-20 on sequential in tiobench, while the other is in the 5-9 range. tiobench on a 2 GB, 16k chunk RAID0 set shows some interesting info: the sequential speeds are about the average of the two disks. the random speeds are about the SUM of the two disks' speeds. As I increase the block size in tiobench, the sequential numbers drop slightly, but the random numbers increase nicely, up to about 3-4 MB/sec with a 32k block size. Q: Is the limp sequential performance due to 1 disk being significantly slower? Q: Any pointers on what a "good" chunk size is, if say, I have many many 4kb files? 2k? Ed -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
Re: stability of software RAID under Linux
Holger Kiehl wrote: > In my opinion Linux SW Raid is very stable. However, SW raid > stresses SCSI to its limit and any tiny error in the hardware As does HW Raid. No matter how you saturate your SCSI bus, its saturated. > will show up very quickly. Also, not all drivers handle > the situation when a disk dies, they hang up the bus. As far > as I know the NCR/Symbios- and AIC7xxx-driver will not > hang up the bus. ECC-Ram is also a must. I wouldn't say that ECC-RAM is a must for RAID systems - if you need it, you need it. -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
help interpret tiobench.pl results?
I get: File Block Num Seq ReadRand Read Seq Write Rand Write DirSize Size Thr Rate (CPU%) Rate (CPU%) Rate (CPU%) Rate (CPU%) --- -- --- --- --- --- --- --- . 20040961 21.57 12.0% 0.634 1.13% 19.67 24.0% 1.080 1.86% . 20040962 16.24 10.1% 0.646 0.84% 19.86 35.7% 1.128 2.67% . 20040964 15.35 9.90% 0.652 0.83% 19.69 36.8% 1.123 2.80% . 20040968 14.82 9.93% 0.671 0.82% 19.56 38.0% 1.126 2.92% The machine only has 64MB of RAM, and it was in X, with Netscape running, so very little memory was free. Seems ok, but the Rand tests seem pretty pitiful. What does this tell me, exactly, esp. as the threads increase? And why does Seq. Read drop off, but Seq. Write doesnt? Ed -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
RAID1 on IDE
If I have a RAID1 set on a single IDE channel, i.e. master & slave, will the box keep running if a drive goes down? -- Edward Schernau,mailto:[EMAIL PROTECTED] Network Architect http://www.schernau.com RC5-64#: 243249 e-gold acct #:131897
RAID and new kernels, FYI
The RAID patches will NOT patch cleanly (nor will much else) on a Redhat-supplied kernel. Make sure you start with a fresh kernel.org tarball. Ed
Install from scratch, RH6.2 + HPT366?
I'm looking for a way to start from scratch - I can install onto the PCI IDE controller (stock), but I need to keep my /dev/hdx entries the same, and LILO to be happy, since I'm dual-booting NTWS, and I dont want to deal with boot issues. I guess its a catch-22, until RH puts support into the boot disks for HPT366? Ed
Re: performance limitations of linux raid
Chris Mauritz wrote: > > Ive done some superficial performance tests using dd, 55MB/s write > > 12MB/s read, interestingly i did get 42MB/s write using just a 2 way ide > > raid0, and got 55MB/s write with one drive per channel on four channels > > (i had no problem writing, just reading) so surprisingly i dont think > > the drive interface is my bottleneck. > > I find those numbers rather hard to believe. I've not yet heard of a > disk (IDE or SCSI) that can reliably dump 22mb/sec which is what your > 2 drive setup implies. Something isn't right. Check http://www.tomshardware.com , they review the Promise IDE RAID card (the hacked one for $30). They get some pretty insane throughputs on some ATA66 drives. > RAID0 seemed to scale rather linearly. I don't think there would be > much of a problem getting over 100mbits/sec on an array of 8-10 ultra2 I tested a 3 drive RAID0 set where 1 disk = 1.7 M/sec, 2 disks = 3.4 M/sec, but 3 disks only = 3.9 M/sec. These were 5 M/sec disks on a AHA2940. I think SCSI bus overhead and contention issues play an important role. Ed -- Edward Schernau http://www.schernau.com Network Architect mailto:[EMAIL PROTECTED] Rational Computing Providence, RI, USA
Re: Fw: Adaptec AAA-133U2 Raid Controller support
Gregory Leblanc wrote: > To add a little more detail, Adaptec has decided that they do not want this > product to work under Linux. The AAA-13X series cards perform ALL raid in > SOFTWARE, using custom drivers that Adaptec has written. I was under the So it's like WinSCSI? Or LoseSCSI? ;-) > P.S. It's not a "windows-only" product, it is also supported under Novell > Netware 4 and 5. Cool, wonder why Adaptec is being so lame about this, and why they'd market a software RAID package as their RAID solution. -- Edward Schernau http://www.schernau.com Network Architect mailto:[EMAIL PROTECTED] Rational Computing Providence, RI, USA
re: Power supply for multiple disks
Mike Bilow wrote: > > Drives use by far more instantaneous current to spin up then they need to > run in active idle mode. A big, modern IDE drive can consume about 2.0A [Excellent analysis trimmed] Nowadays I think the "danger" comes from two sources. First is poor PS quality, and the second is the increased power demands of other components. Cheap PS' will not be able to hold the voltage constant under an instantaneous load. Even the best can't, but its a measure of how "stiff" the PS is if it can maintain a stable +5V line, under sudden load. Dips on this line (and others) can affect the entire system. Secondly, processors and video cards can suck power. I have 2 overclocked Celerons running at 522 MHz, EACH with its own Alpha cooler + 60x20mm fan. I have an 80mm case fan in the bottom of the case, with a 60x20mm fan mounted in the top. Completing the package is a Voodoo3 card with a "slot exhauster" fan. I also have a 100BaseT NIC, internal ISA modem and vintage SB16, plus an AHA2940 for my CDRW (internal). I'm using the 235W ATX PS that came with the case. Where you _can_ run into trouble is with a PS loaded to a good percentage of its capacity, THEN hit with an instantaneous load. Suddenly your +5V line can bounce around, which can cause memory errors, CPU errors, etc. -- Edward Schernau http://www.schernau.com Network Architect mailto:[EMAIL PROTECTED] Rational Computing Providence, RI, USA
re: aic7xxx, SMP, "providence" board
How do you get interrupts 17 and 18 ??? -- Edward Schernau http://www.schernau.com Network Architect mailto:[EMAIL PROTECTED] Rational Computing Providence, RI, USA
purpose of --really-force
Is it JUST to attract flames? Since it breaks when you dont have the right kernel, shouldn't it simply return a "dude, get a kernel" message, instead of trying and failing? -- Edward Schernau http://www.schernau.com Network Architect mailto:[EMAIL PROTECTED] Rational Computing Providence, RI, USA
2.3.x ??
Is raid 0.90 in the 2.3 kernels, or do you still need to patch? Sorry if this has been discussed before... -- Edward Schernau http://www.schernau.com Network Architect mailto:[EMAIL PROTECTED] Rational Computing Providence, RI, USA
network RAID
for the Win32 equivalent, check out mango medley (forget URL) -- Edward Schernau http://www.schernau.com Network Architect mailto:[EMAIL PROTECTED] Rational Computing Providence, RI, USA
Re: Raid over Network
Murat Koc wrote: > > Hi all, > I have 40 comps and have 2GB free partition each of them. > I want to manage them on one comp in one directory. > When I asked this question to linux-admin list, somebody told that there > is a solution on this list with subject raid over network. > If there is something like that, could anybody explain how I can do that? > Or recommend any doc. You can use AFS, a network file system, or _I THINK_ something with the Network Block Device. Check in the kernel docs. -- Edward Schernau http://www.schernau.com Network Architect mailto:[EMAIL PROTECTED] Rational Computing Providence, RI, USA
crash/instability
Note the crosspost. Ive had occasional, maybe 6 lockups since I bought my ABit BP6/dual system, a month ago. Since installing the raid patches on my hpt366-patched kernel last week, its locked up much more often, tonight 4 times. I left it running all day with 2 RC5 clients running, and it was fine. Within 2 hours of starting X, loading WP8, netscape, etc., it died 3 times. This seems more frequent since I patched for RAID. Comments? -- Edward Schernau http://www.schernau.com Network Architect mailto:[EMAIL PROTECTED] Rational Computing Providence, RI, USA, Earth
re: kernel patch
Make sure you get a real kernel to start with from kernel.org, not a kernel-o-matic.rpm from RH. -- Edward Schernau http://www.schernau.com Network Architect mailto:[EMAIL PROTECTED] Rational Computing Providence, RI, USA, Earth
RAID0 works, RAID5 dies, why?
On the same 3 SCSI disks, I can create a RAID0 set, and run repeated bonnies on it. On these disks, if I then make a RAID5 set, it starts making the /dev/md0, and then dies, flooding my console with scsi bus reset error messages. The higher the chunksize I use, the later this happens, but I can't get RAID5 to work. It's a dual celeron 550 box, 2.2.13 patched, AHA2940, and 3 SCSI2 (but 5 MB/sec) 420 MB Seagate disks. Something about SCSI bus activity, either breaking the drives or the card driver? Any ideas? I have some interesting bonnie runs, if people care. -- Edward Schernau http://www.schernau.com Network Architect mailto:[EMAIL PROTECTED] Rational Computing Providence, RI, USA, Earth
RAID0 problem
Note the crosspost. patching with the ALPHA kernel patch and recompiling gives much more verbose, but equally unhelpful messages about WHY my RAID-0 array fails. I have raidtools-0.90, and the most recent RAID kernel patch. Supposedly 2.2 was supposed to handle this but I guess that's a lie. Does anyone have this working? -- Edward Schernau http://www.schernau.com Network Architect mailto:[EMAIL PROTECTED] Rational Computing Providence, RI, USA, Earth
kernel patch?
I am running 2.2.13, whose config script has options for RAID. I have raidtools-0.90. Why/Do I need to patch? Pointers appreciated. -- Edward Schernau http://www.schernau.com Network Architect mailto:[EMAIL PROTECTED] Rational Computing Providence, RI, USA, Earth
test, and a question
Is this list active? If so, I have a question Running 2.2.13 with patches for the HPT366 IDE controller. raidtools-0.90 AHA2940, 2 411 MB SCSI drives, sda and sdb. I've made a single partition on each, not yet formatted. I'm trying to create RAID0. My /etc/raidtab is: # Sample raid-0 configuration raiddev /dev/md0 raid-level 0# it's not obvious but this *must* be # right after raiddev persistent-superblock 0# set this to 1 if you want autostart, # BUT SETTING TO 1 WILL DESTROY PREVIOUS # CONTENTS if this is a RAID0 array created # by older raidtools (0.40-0.51) or mdtools! chunk-size 16 nr-raid-disks 2 nr-spare-disks 0 device /dev/sda1 raid-disk 0 device /dev/sdb1 raid-disk 1 Totally standard, I think. When I run "mkraid /dev/md0" I get: handling MD device /dev/md0 analyzing super-block disk 0: /dev/sda1, 420848kB, raid superblock at 420736kB disk 1: /dev/sdb1, 420848kB, raid superblock at 420736kB mkraid: aborted What does this mean? Thanks in advance. -- Edward Schernau http://www.schernau.com Network Architect mailto:[EMAIL PROTECTED] Rational Computing Providence, RI, USA, Earth