Re: Installing Fedora Core with root on Reiserfs
On Thursday 21 July 2005 10:04, Edward Shishkin [EMAIL PROTECTED] wrote: My phrase reiserfs.ko located on reiserfs sounds bad, and I should clarify that the reiserfs.ko is contained in the initrd with other binaries/scripts, and this initrd looks fine from the standpoint of kernel/reiserfs, but not from the standpoint of grub/reiserfs-emulation. The logs obtained from serial console don't include anything about loading initrd, and there is the following detail: a dump created by debugreiserfs -d shows that the initrd (i_size: 1128235) is represented by an indirect item (276 4K-blocks), while grub found that this is not sector-aligned: Might this be related to the size of the ReiserFS file system? I tested installs with the default partitioning (100M /boot) which worked OK. When you had the problem were you using a larger ReiserFS file system? -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
Re: Installing Fedora Core with root on Reiserfs
On Tuesday 19 July 2005 01:59, Jeff Mahoney [EMAIL PROTECTED] wrote: If the root file system is reiserfs then reiserfs.ko will (or at least should) be included in the initrd. Right, but initrd is in /boot which is not something separate: it is on the same reiserfs root partition.. The situation you're describing is one that is well tested by now. If the root filesystem is reiserfs, and /boot is a part of it, reiserfs.ko MUST be in the initrd. Otherwise, there is a chicken/egg problem and the system will not boot. This works in all my tests. The reiserfs.ko module is apparently in the initrd. Also if the original bug concerned a lack of reiserfs.ko in the initrd then re-running GRUB would not fix things. I can't reproduce the bug, it just works for me. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
Re: Installing Fedora Core with root on Reiserfs
On Monday 18 July 2005 06:01, Edward Shishkin [EMAIL PROTECTED] wrote: FC4-test3 (and perhaps FC4) installs its own version of grub which seems to interact incorrectly with reiserfs. The problem is that reiserfs.ko module located on reiserfs partition can not be loaded. Firstly there is no situation in which reiserfs.ko will be loaded from a reiserfs partition. If the root file system is reiserfs then reiserfs.ko will (or at least should) be included in the initrd. The GRUB support for ReiserFS is based on the file /boot/grub/reiserfs_stage1_5 which is only needed for /boot on ReiserFS. Some people believe that /boot should be a separate file system to the root file system regardless of all other issues. All systems which use LVM for the root file system on Fedora (the default partitioning involves LVM) will have a separate /boot file system because GRUB apparently doesn't support LVM (yet). So having a /boot with Ext2/3 is an easy solution to any ReiserFS booting issues, and a solution that will be forced on the users if they use LVM. When using a small /boot file system ReiserFS is not a good choice. A quick test showed that Ext3 gave 18M more usable disk space from a 100M file system than ReiserFS, this will essentially force all LVM users to use Ext3 for /boot. 1. (Requires some partition formatted by ext2). At the end of installation process (after invitation to reboot) boot from some rescue CD, mount reiserfs root partition (say to /mnt) and move all the compressed images (vmlinuz-xxx and initrd-xxx located at /mnt/boot) to your ext2 partition. Then edit the file /mnt/boot/grub/grub.conf pointing a new location for the images. Reboot and finish the installation process. This can be done automatically through the Anaconda GUI or through kickstart. Just specify that /boot is to be a separate partition of type Ext3. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
Re: Write-once file system
On Sat, 28 Jun 2003 03:09, Jason Holt wrote: On Fri, 27 Jun 2003, Fong Vang wrote: I don't think turning the write option off during write is a good idea. All file systems running reiserfs should make the file write-once. File systems that do need to be rewriteable will use ext3 or something else (that's how we do it now anyway). Could it done in such a way that even root can't write (not even when using block devices)? [...] The trick is that root controls the kernel, and the kernel talks directly to the hardware. That's all a block device is - (mostly) direct hardware access. So what you're asking for is something beyond root's control that can tell him no when he asks to write to an immutable file. Another option is to use a security system such as SE Linux to limit the access given to the root account. In SE Linux a daemon running as root generally has very little access to the system, and a UID=0 user who is in the user_t domain gets less access than a non-root user on a non-SE system. Go to my SE Linux web page (below) and read about my play machine. SE Linux works well on ReiserFS. I don't use ReiserFS on my play machine however because it can only boot from Ext2, Ext3, or XFS. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
Re: xattr
On the topic of atomic xattr operations on ReiserFS as needed for the new LSM/SE Linux operations. On Thu, 19 Jun 2003 23:52, Chris Mason wrote: How big are the xattrs you have in mind? We can get atomic writes of 4k in length but beyond that things get more difficult. Most of them will be less than 80 bytes. They are currently of the form: user-name:object_r:type The user-name is the Unix account name which usually isn't much more than 8 bytes. The type is usually less than 15 bytes (the longest I've used so far is 20 bytes). So the longest value I've used is 38 bytes. Also they can't be chosen arbitarily by the user. The user gets some small control over the type within a range of types that the administrator permits. If the administrator permits overly long type names and has to deal with non-atomicity as a result then it's their issue. If you can guarantee atomic operations on 160 byte operations (twice what I expect anyone to use) then it'll be fine. As for the xattr and the create in the same transaction, that's a little harder. We'd probably need a new syscall, or to change the semantics of the xattr call such that creating an xattr on a file that doesn't exist also creates the file. Creating a file by creating the xattr sounds like a bad idea as you can't control the Unix permissions of the file. This isn't much of a big deal with SE Linux as the security type determines who can access the file. But for other uses it may be a serious problem. I agree that we need a new syscall and other people had the same idea before either of us. Maybe ReiserFS could be used as the first implementation of this proposed new syscall... -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
Re: xattr
On Fri, 20 Jun 2003 00:55, Chris Mason wrote: So the longest value I've used is 38 bytes. Then data=journal mode will do what you want. You'll get atomic writes up to 4k. If you really don't want data=journal for the rest of the FS, we can make an option for using data logging on xattr files only. Jeff and I had wanted to avoid the complication but it is at least possible. OK. Initially just using data=journal should be fine, and even for wide-spread use, forcing everyone to use data=journal shouldn't be too much of a hardship - although I'm sure that some people will prefer to journal only the xattr's. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
xattr
What is the status of xattr support in 2.5.x? How is journalling of xattr's being handled? For correct operation of SE Linux we need to have the xattr that is used for the security context be changed atomically, and if a file is created and immediately has the xattr set then ideally we would have the file creation and the xattr creation in the same journal entry. Is this possible? If doing this requires that the file system be mounted with data=journal then this will be fine. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
another IDE-DISK issue in 2.4.21-rc3
I have finally installed 2.4.21-rc3. I compiled the base IDE driver into the kernel but make the ide-disk driver a module (to save RAM on machines without IDE hardware). Following is a selection from the kernel message log. As you can see I get some nasty and worrying error messages when the kernel tries to access /dev/hda before the ide-disk module is loaded. If this is not a known bug then let me know so I can send it to linux-kernel. ReiserFS seems to be working fine in 2.4.21-rc3. Uniform Multi-Platform E-IDE driver Revision: 7.00beta3-.2.4 ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx PIIX4: IDE controller at PCI slot 00:07.1 PIIX4: chipset revision 1 PIIX4: not 100% native mode: will probe irqs later ide0: BM-DMA at 0x1c10-0x1c17, BIOS settings: hda:DMA, hdb:pio ide1: BM-DMA at 0x1c18-0x1c1f, BIOS settings: hdc:pio, hdd:pio hda: IC25N030ATCS04-0, ATA DISK drive blk: queue c02a8680, I/O limit 4095Mb (mask 0x) ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 Partition check: hda:end_request: I/O error, dev 03:00 (hda), sector 0 end_request: I/O error, dev 03:00 (hda), sector 2 end_request: I/O error, dev 03:00 (hda), sector 4 end_request: I/O error, dev 03:00 (hda), sector 6 end_request: I/O error, dev 03:00 (hda), sector 0 end_request: I/O error, dev 03:00 (hda), sector 2 end_request: I/O error, dev 03:00 (hda), sector 4 end_request: I/O error, dev 03:00 (hda), sector 6 unable to read partition table [boring stuff removed] hda: attached ide-disk driver. hda: host protected area = 1 hda: 58605120 sectors (30006 MB) w/1768KiB Cache, CHS=3876/240/63, UDMA(33) /dev/ide/host0/bus0/target0/lun0: p1 p2 p4 p5 p6 p7 p8 -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
Re: Multiple data streams...
On Wed, 2 Apr 2003 19:28, Anders Widman wrote: Is this supported, or will it be supported by ReiserFS? I use this feature quite quite much.. Maybe this is something to add to ReiserFS? There is very brief info at Microsoft's website: Does MS support multiple data streams properly yet? The last information I read about data streams indicated that commonly used operations (such as typing copy in a command-line session) would not work properly with multiple data streams. Also many important programs such as anti-virus programs did not support it. The Linux API does not support multiple data streams. Support for extended attributes has recently been added, this will do most things that you might want to do with multiple streams and does it in a standard way that is supported. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
Re: filesystem corruption ?
On Fri, 21 Mar 2003 14:07, Oleg Drokin wrote: I've learn in the school that if you put some bit amount of plumbum in It's better known in English as lead. The problem with lead is that it's poisonous and soft. Having to wash your hands after touching your computer could get annoying. Other metals such as copper and steel will reduce the radiation and can also be used for protection against mechanical damage. The best way to reduce radiation is by distance. The inverse-square law applies, so moving the computer further away from the experiment will reduce the radiation more easily than anything else you may do. One thing to consider is disk-less X-term machines for if you need to operate a computer from near the experiment, so if the X-term crashed from radiation then your server with the data should continue running correctly. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
Re: Corrupted/unreadable journal: reiser vs. ext3
On Wed, 12 Feb 2003 16:26, Anders Widman wrote: Unplanned downtime do cause lot of harm to any business. It's better to stop when there's a serious error than to blindly continue and make things worse. I (and I think no one else) never said continue blindly. Most users/workstations do not have RAID and probably never will. Hard drive costs are constantly decreasing while the value of data is constantly increasing. I think that the use of RAID will increase steadily. The others want to make Linux a viable option for normal users and want Linux to be able to replace Windows or Mac OS. The only way I see that happen is if Linux starts to get more userfriendly and safe. I guess you're not familiar with what NT does then. NT 3.5x would sometimes get confused about it's data and umount the file system in question to avoid the risk of damaging data. In case of a serious kernel error NT will give a BSOD in situations where Linux by default will print an Oops message and continue running. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
Re: OT: Swapfile to RAM relation with 2.4.2X+
On Fri, 7 Feb 2003 02:47, Manuel Krause wrote: In the beginning of 2.4.0+ a relation of swapfile-to-RAM of 2-to-1 was recommended. Due to my several system changes to come in those times I Such recommendations are only generalisations. Ignore them and look at what your system is doing. If your swap space never runs out and you don't expect your usage patterns to require more (including cron jobs and periods of unexpected load) then you have enough. If you run out of swap space then you need more, also you should have some swap even if you have a lot of memory. There's always data that isn't used much and can be paged out to make room for more disk cache. BTW Anything that is worth saying in a .sig can be said in 4 lines. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
Re: kernel go-slow
On Thu, 6 Feb 2003 17:32, Alexander Lyamin wrote: One problem that has started occuring is that periodically some of the machines will go really slow for a while. It's as if the CPU speed has just dropped to 1% of it's regular speed. Then after 10 minutes or so it will continue as normal. when its slows down, please check with vmstat for IO or with your i think i wasnt clear enough. so - first , if you go-slow on a disk activity, chances are good that it caused by FS or VM or their misunderstandings. vmstat doesn't work properly. CPU time is 99% system which suggests that one CPU is spending all it's time in kernel space (for both threads of a hyper-threaded CPU) or that both CPUs have each got one thread locked in kernel space. It's not disk related, those machines don't have a huge disk access. The machines with the serious disk activity don't have any problems. but there is possible situations that will not generate disk activity, but may cause your system to go-slow, if there you have some unussual IO numbers while disk activity is moderate to low - most likely same sweet pair. The problem is that sar etc product jumbled results. Profiling the kernel may help, but may also hide the error, and it's not something I can easily do. The servers are locked in a managed server room on the other side of the city so seeing the blinken lights is not an option. I've put the aa1 kernel on half the machines and now I'll wait to see what happens. If the aa1 machines don't have the problem but the others do then I'll go all aa1. Interestingly the machines that have the problems are not the most active in the file system (mail store), but the mail spool machines. The mail spool machines do a good amount of file access (but well below the limits of the hardware) and also use more memory and have large load spikes on occasion (virus and spam scanning). talking about virus/spam scanning - what do you use and how its integrated in your SMTP MTA ? RAV. I'm not sure of the details, I think it runs as a daemon that qmail talks to. I try to avoid the anti-virus stuff. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
kernel go-slow
I'm running a number of machines with 2.4.20 and the ReiserFS journal patches. One problem that has started occuring is that periodically some of the machines will go really slow for a while. It's as if the CPU speed has just dropped to 1% of it's regular speed. Then after 10 minutes or so it will continue as normal. Has anyone heard of such things before? I am asking here first because the ReiserFS patch is the most significant kernel patch I've applied on what is otherwise a stock 2.4.20 kernel. Interestingly the machines that have the problems are not the most active in the file system (mail store), but the mail spool machines. The mail spool machines do a good amount of file access (but well below the limits of the hardware) and also use more memory and have large load spikes on occasion (virus and spam scanning). -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
kswapd CPU usage and heavy disk IO
I have a server with 4G of RAM running ReiserFS for everything that matters. It has 2G of swap space free, but so far I have not seen swap usage go above 1.6M (so in normal use I could turn off swap entirely and expect not to see much difference). When it's under really heavy load (when I have a maintenance task involving a find / and there are lots of POP/IMAP clients hitting the server as well as mail delivery) and the load average gets to about 40, the kswapd kernel thread starts using excessive CPU time. It will stay on ~4% but have spikes of up to 45%!!! This is a two-processor machine so 45% CPU reported by top means 90% of a single CPU I guess. 90% of a 1.8GHz P4 CPU is a lot of CPU and I think that something is wrong. In the meager documentation in the kernel source kswapd is described as being involved in paging to disk. I don't think that this is what it is doing as there is no noticable paging activity (it generally has at least 600M of buffers so there is no real shortage of memory). Would the activity of kswapd be involved with ReiserFS in any way? What can I do to improve this situation? -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
Re: what do you do that stresses your filesystem?
On Mon, 23 Dec 2002 16:56, Ross Vandegrift wrote: On machines with Linux md RAID arrays that need to be remirrored, I do end up waiting a bit (10-30 seconds) due to read starvation, but this problem is so much better than it used to be it's hardly worth mentioning. One thing I've done before is configure the boot scripts to set the RAID re-sync speed very low before fsck/mount time and then set it back later. It didn't seem to do as much good as I thought it would though. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
non volatile ram devices
I have some servers that are giving inadequate disk performance for Maildir mail spools. They are running kernel 2.4.19 (2.4.20 upgrade is planned) and using ReiserFS for everything that's important. At this stage it is impossible for me to replace disks, RAID controllers, or anything else really significant. What I am thinking of doing is using a kernel that supports data journalling which should increase performance, but still probably won't give me enough. So I am thinking of using an external journal (or using software RAID to put the part of the partition containing the journal on a different device). The device containing the journal would be something much faster than physical media. I have been doing some research on non-volatile memory devices. I only found one company producing disks that are RAM based with battery backup, and they seem to start at $10K (too expensive - probably because they are much larger than I need, I need 128M at most, they provide 2G). I found many companies selling flash memory, but that only takes a million writes (that'll last about an hour for the use I plan). I found one company selling PC-Card devices that have two batterys for backup, but that requires getting a PCI controller for PC-Card's (something I haven't tried before). Does anyone know of an affordable ($1000 or less) device that can survive unexpected power outages of at least 24 hours duration, can commit a write in less than 1ms, supports unlimited writes, and connects to a IDE or SCSI bus (or PCI if there's a suitable Linux driver). -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
2.4.20 and data logging
How is data logging work going? Is there a patch for 2.4.20? -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
Re: other OS
On Tue, 3 Dec 2002 05:11, darren wrote: Was just wondering about the possibilities of using reiserfs on my Solaris 6 machine. Any chance of that? I doubt it. I can't imagine Sun (or anyone else) paying Hans enough money to do this. Veritas VXFS is good for this type of thing. Any time you want half decent IO performance on Solaris you want Veritas VXFS and Volume Manager. I am having problems with FS performance (large number, 20K of small files in folder with constant writes and deletes) on the Sun Machine and was thinking if this can help me. The only way I know is to use a Linux machine running reiser and NFS it over. But performance will surely be degraded this wayany other suggestions? In 1999 I was doing some work on AIX machines. I found that an AIX server was so slow that NFS mounting a ReiserFS file system over a 10baseT network was faster for some operations!!! If you do a usenet search you can find references to this, comp.sys.aix or something. Another option is to use Linux on the SPARC machine. Linux on SPARC reputedly performs a lot better than Solaris if you have 8 or less CPUs, even without the ReiserFS issue. I think that Linux on SPARC can even run Solaris binaries (but I'm not certain). PS There is no such thing as Solaris 6. They jumped straight from Solaris 2.6 (which is presumably what you are using) to Solaris 7.0. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] Reiserfs with Samba vs NetApp Filer
On Tue, 15 Oct 2002 15:42, Hans Reiser wrote: Russell Coker wrote: See the following graph: http://www.coker.com.au/~russell/hardware/46g.png This shows testing a single 46G drive, two drives on different buses at the same time, and two drives on the same bus at the same time. zcav (part of Bonnie++) was used to perform the tests. I am surprised that separating them onto different buses has so little effect. It looks like most of the bottleneck for large reads off two drives is not the IDE bus, but something else (maybe CPU or memory bandwidth). I was surprised too. Especially as it's an ATA-66 bus (the bus was expected to be a bottleneck). Only a single CPU. I would like to do more research on this matter and write a magazine article (I already have a magazine wanting to publish it). All I need is suitable access to the latest hardware to perform my tests (tests on old hardware while still being interesting research doesn't sell magazines). -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
[reiserfs-list] back up to disk
Here's an interesting article I just read. It's just a device with a bunch of ATA drives inside, up to 2T of storage. Probably anyone here could produce something based on ReiserFS to compete with it... Storage start-up Avamar Technologies is launching an appliance this week that it claims backs up network data more quickly and less expensively than tape. http://www.nwfusion.com/news/2002/1014avamar.html?net -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
[reiserfs-list] This list is good for spammers.
There has been a huge amount of spam coming through this list recently. Could the list master please take the following actions: 1) Make the list server software not strip all headers so we can see how the messages get to the list server? Then we could report spam to SpamCop, and people with good filtering software could filter on the first IP address in the list. 2) Make the list partially moderated. Suggestion, moderate any message from someone who is not subscribed, and any message from an IP address that's in the common DNSBL's. This will not deny anyone access to the list and will not take a huge amount of effort (once the few people like myself who subscribe with one address and post with another get with it). Doing such things doesn't seem to be difficult, everyone else is doing it. Another more invasive action is to make the list server use some of the DNSBL services to reject postings altogether. It will slightly inconveniance some legit users, but will really cut down on spam. -- I do not get viruses because I do not use MS software. If you use Outlook then please do not put my email address in your address-book so that WHEN you get a virus it won't use my address in the From field.
Re: [reiserfs-list] fsync() Performance Issue
On Fri, 26 Apr 2002 22:28, [EMAIL PROTECTED] wrote: It's interesting to note your email address and what it implies... I'm wondering if anyone out there may have some suggestions on how to improve the performance of a system employing fsync(). I have to be able to guaranty that every write to my fileserver is on disk when the client has passed it to the server. Therefore, I have disabled write cache on the disk and issue an fsync() per file. I'm running 2.4.19-pre7, reiserfs 3.6.25, without additional patches. I have seen some discussions out here about various other speed-up patches and am wondering if I need to add these to 2.4.19-pre7? And what they are and where can I obtain said patches? Also, I'm wondering if there is another solution to syncing the data that is faster than fsync(). Testing, thusfar, has shown a large disparity between running with and without sync.Another idea is to explore another filesystem, but I'm not exactly excited by the other journaling filesystems out there at this time. All ideas will be greatly appreciated. These issues have been discussed a few times, but not with any results as exciting as you might hope for. One which was mentioned was using fdatasync() instead of fsync(). One thing that has occurred to me (which has not been previously discussed as far as I recall) is the possibility for using sync() instead of fsync() if you can accumulate a number of files (and therefore replace many fsync()'s with one sync() ). -- If you send email to me or to a mailing list that I use which has 4 lines of legalistic junk at the end then you are specifically authorizing me to do whatever I wish with the message and all other messages from your domain, by posting the message you agree that your long legalistic sig is void.
Re: [reiserfs-list] Encryption plugin developer needed for reiser4
On Tue, 12 Mar 2002 19:43, Sam Vilain wrote: I've done some benchmarking of the old international kernel patch and found it to be usable on small systems. Done it recently? hoffman:~$ df . crypto/ Filesystem 1k-blocks Used Available Use% Mounted on /dev/hda7 6265120 5366380898740 86% /home /home/sam/.crypto 665572498796166776 75% /home/sam/crypto hoffman:~$ time bash -c dd if=/dev/zero of=test bs=4096 count=10240; sync 10240+0 records in 10240+0 records out real0m5.152s user0m0.050s sys 0m0.640s hoffman:~$ time bash -c dd if=/dev/zero of=crypto/test bs=4096 count=10240; sync 10240+0 records in 10240+0 records out real0m9.685s user0m0.030s sys 0m0.660s hoffman:~$ So we're doing 40M in 10s, this means something like 12MB/s encryption speed. That's with AES, a 192 bit key size, and a 2.4.18-ac3+preempt kernel. My machine is a fairly new Dell(850MHz). Both filesystems are reiserfs. During a fsck, the CPU isn't doing much else anyway. So it would take about twice as long, assuming you don't have a system more powerful than my laptop to do the encryption. Say you had a dual processor system (or a single Athlon ;), it could probably encrypt/decrypt as quickly as the disk can transfer data, especially for random access. If a fast Athlon is twice as fast then it'll still be a bottleneck if you have a single fast IDE hard drive (modern IDE drives can sustain 30M/s for linear transfers). Then think if you have 10 large file systems each comprised of 6 disks in a hardware RAID array. You'll never get enough CPU power to keep up. -- If you send email to me or to a mailing list that I use which has 4 lines of legalistic junk at the end then you are specifically authorizing me to do whatever I wish with the message and all other messages from your domain, by posting the message you agree that your long legalistic sig is void.
Re: [reiserfs-list] O/T but expert answer needed: MS says NTFS does full data journaling
On Thu, 14 Feb 2002 20:25, Paul Robertson wrote: When a maching gets an Oops or BSOD condition then the kernel is inherantly doing improper and unpredictable things with memory. Therefore regardless of what file system you use it could get trashed and data could get lost. Oops conditions are generally rare on Linux machines so this shouldn't be much of an issue. BSOD on NT is quite common... IMO oops and BSOD are quite different. There are many possible reasons why an NT kernel component might decide to call KeBugCheck() which generates the BSOD. I have a book which lists around 100 common bugcheck codes. In particular, NT can be configured to dump the system state to a file on the boot partition when a crash occurs. There are also a couple of Linux kernel patches to support dumping the memory to the swap partition on an Oops, and an Oops can be triggered by any condition that some kernel code considers Oops-worthy. IMHO The biggest difference between an Oops and a BSOD is that a machine doesn't totally die after an Oops (which can be considered a good or a bad thing). -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] O/T but expert answer needed: MS says NTFS does full data journaling
On Thu, 14 Feb 2002 02:58, [EMAIL PROTECTED] wrote: On Wed, 13 Feb 2002 12:26:59 +1300, Adam Warner [EMAIL PROTECTED] said: Does Windows journal the metadata, data or both? Answer: Windows NT/2000 systems that utilize NTFS since NT3.1 have always journalled and logged metadata and data, so we've been doing this for close to a decade. I just want to confirm if this is in fact true. I can't find a Hint: If they journal both, why do you ever hear of people getting corrupted filesystems when the box BSOD's? (No, I don't know if it does or not - but I've heard *too* many people say It hosed the disk and I had to reinstall for me to think that it's done correctly) When a maching gets an Oops or BSOD condition then the kernel is inherantly doing improper and unpredictable things with memory. Therefore regardless of what file system you use it could get trashed and data could get lost. Oops conditions are generally rare on Linux machines so this shouldn't be much of an issue. BSOD on NT is quite common... -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] Can ReiserFS handle removable storage?
On Mon, 21 Jan 2002 07:28, Oleg Drokin wrote: USB by it's nature is something external to the system. Unplugging a USB cable with a mounted drive attached should (IMHO) get the same result as unplugging an Ethernet cable with an NFS mount in progress. This means processes go into D state if they have outstanding writes, and for reads they may go D state depending on mount options, and then you wait for the device to become available again. How do you distinguish between SCSI USB storage in Linux on fs level? ;) You can have SCSI and IDE unpluggable devices too... For a file system on USB ReiserFS would have to recheck the superblock (to make sure that it hasn't been mounted on another computer in the mean time) before allowing access again. Also there would have to be a recovery process for the situation when the USB device is gone for good. Sound not very easy to do ;) True. Writing a good file system is never easy. -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
[reiserfs-list] transient hard drive error causing problems
Below is the relevant portion of my dmesg output. I get a status error on my first hard drive (it's a transient thing after a hard reset). There are a few issues here: The dmesg output does not tell me which partition the error refers to! I can presume that it's related to the error on hda, but as hda has parts of several RAID devices even that doesn't narrow it down much. Can the /proc/reiserfs stuff be used to track this down? Also there's the usual issue of messages having codes vs-number, is there a reference to what the numbers mean? hda: status error: status=0x58 { DriveReady SeekComplete DataRequest } hda: drive not ready for command is_tree_node: node level 759 does not match to the expected one 2 vs-5150: search_by_key: invalid format found in block 16618. Fsck? vs-13070: reiserfs_read_inode2: i/o failure occurred trying to find stat data of [13396 11199 0x0 SD] hdb: ATAPI 40X DVD-ROM drive, 512kB Cache, UDMA(33) Uniform CD-ROM driver Revision: 3.12 is_tree_node: node level 759 does not match to the expected one 2 vs-5150: search_by_key: invalid format found in block 16618. Fsck? vs-13070: reiserfs_read_inode2: i/o failure occurred trying to find stat data of [13721 31689 0x0 SD] is_tree_node: node level 759 does not match to the expected one 2 vs-5150: search_by_key: invalid format found in block 16618. Fsck? vs-13070: reiserfs_read_inode2: i/o failure occurred trying to find stat data of [13396 14810 0x0 SD] -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] When will Reiserfs be ready?
On Mon, 7 Jan 2002 11:48, _nasturtium wrote: What has happened to NTFS performance? During the Windows NT 3.5 days it was the slowest FS in production use, particularly for small files. Has this changed? (Should I go from FAT32 to NTFS on my laptop windows partition? Fear of performance loss has kept me from doing so.) For an architect of a filesystem, you aren't giving it much support. You are very intent on parading your stupidity in this list. This list is (hopefully) to share knowledge and solve problems, not to try and insult people. You've changed your mind rapidly on this issue. What are you doing having a FAT32 partition??!!! Running Windows as he clearly states in his message. While we might debate the issue of whether he should be running Windows or whether he should use Star Office or other software when dealing with people who use Word documents, we can't debate the suitability of FAT32 for Windows. Have a minimal FAT32 partition, then run reiser4win as discussed by Gerson Kurz, Yves Glodz and friends. Then you can run windows (for what purpose??!!) and develop reiser simultaneously. Which was not what you suggested before. Also it wouldn't work anyway as Linux file systems are well integrated into the kernel buffering (this is either a good thing or a bad thing depending on your opinion). So porting work from Windows to Linux would involve adding a lot of new buffering code and testing it. I suspect that porting from Linux to Windows would be easier (porting from Linux to OS/2 would definately be easier than porting from OS/2 to Linux - and I suspect that Windows still has some similarities with OS/2 in this regard). Although if you must use windows, DON'T swap to NTFS because Linux support is read-only. Support for writing to NTFS has been working for several years. I was doing it in 1999. There are issues about it, and the code is still marked experimental, but it works. Hmmmwhy not join the Linux-Mandrake Newbie list at www.mandrakesoft.com? You might learn a fair bit - civileme [EMAIL PROTECTED] who is, or at least was on the Mandrake QA team comments on the Help! WinXP thread (4/1/02, 22:20) that: But FYI the XP NTFS is different from the NTFS5 W2k NTFS which is different from previous NTFS setups. We can read and experimentally write NTFS for winNT4, but not NTFS5 for Win2k nor the WinXP version of NTFS. Microsoft has made the format a trade secret. The important part is you CANNOT read/write for Win2k NTFS...if you want to call him an idiot go ahead...join the list, I'm on it - that post was in reply to one of my mails... You initially said that NTFS support in Linux is read-only. Now you say that Windows XP NTFS support is read-only, which is totally different. I said that Linux has support for writing NTFS and that I had used such support in 1999 (long before the release of Windows XP). -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] When will Reiserfs be ready?
On Mon, 7 Jan 2002 12:15, _nasturtium wrote: I was reading the FAQ on www.namesys.com and it seems Reiser4 is sponsored (but not endorsed by...) by DARPA (Defence Advanced Research Projects Agency?). That seems like a good source of funds compared to your support business - your Support page only claims one request/payment every few days. You should try running a business when you grow up. Then you will discover the perilous position that having only a single customer can place you in. Looks like my post was a bit incomplete - Reiser is also sponsored by SuSE (formerly mp3.com), ApplianceWare and BigStorage Inc. You can verify it by going to www.reiserfs.com. Having 4 probably contracted sponsors is a good way to have a business. Read The Code Book (Simon Singh) or any good history book and you will find that the Enigma cipher machine was bought almost totally by the German military. Scherbius, the inventor, made a LOT of money. One example does not prove the generic case. For every inventor who makes a lot of money there are at least 100 who don't. Most new businesses don't last 2 years, most small businesses don't last 10 years. If you run a business and want it to last you have to have contingency plans and backups. Tieing everything to one source of income is very dangerous for a small business, especially if what you are doing is not a core business area for your customer. ReiserFS isn't important enough for the US government to take such a risk. Also Hans just mentioned that Applianceware went out of business, surely that demonstrates how it is better to have multiple sources of income. There is only one use of the user-pays support every few day - someone commented it was because there are hardly any errors. If questions can be answered on this list, why would anyone pay??!! Hans Reiser is too helpful on this list - if he didn't answer namesys would make a lot more money. So now you're saying that Hans is too helpful for his own good and that he should cease providing free help? What do you mean by when you grow up? Perhaps you should resit the German history test - if you're even a freshman. What is the relevance of German history to ReiserFS discussion? -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] magic is useless Determining File Types
On Mon, 7 Jan 2002 10:49, Hans Reiser wrote: There is an issue of going completly overboard, attribute/subattribute/subsubattribute anybody? This is certainly an overall interesting idea. How about file//acl for accessing ACLs? This does mean though you *MUST* have a filesystem specific dump tool. Yep, we have to improve tar. Also we must not break the tar file format!!! Please keep in mind my previous messages on this list regarding LHArc and OS/2's EAs when thinking of changes to tar. The big advantage of tar is that it's files can be read on any OS so no matter how much hardware and software I lose then I can still find a way to read my tar files! -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] When will Reiserfs be ready?
On Mon, 7 Jan 2002 15:14, Andre Pang wrote: Some people need to run Windows to run various applications. It has many apps available which Linux does not; accept it. If Absolutely. In this case if you are doing sales then you MUST have MS software. There are enough hassles in applying for a job without using MS software. I simply refuse to send my CV in any format other than HTML. Any recruiting agent who can't work out how to make MS-Word import HTML or to make IE load it isn't qualified to read my CV or to represent me. I've had a few arguements about this issue with recruiting agents. I suggest that it's best of Hans uses the default options for every other program he uses so that he can concentrate his energies on ReiserFS (but I'm sure he's already doing that). Then why are you suggesting that Hans uses Explore2fs and reiser4win on WinNT? They're certainly not defaults. You misread the quoting. That's something I wrote. Support for writing to NTFS has been working for several years. I was doing it in 1999. There are issues about it, and the code is still marked experimental, but it works. [Russell: where it works meaning usually works but may bugger up your filesystem badly ;)] Yes, they always had warnings about that. So I just made some good backups and gave it a go. I never lost any data (and what I was doing was simple enough that there was no chance of losing data and not realising it). But I admit that I didn't overly stress it. Anyway in that case I wouldn't have minded saying oops I trashed that NT machine and lost the CD - I'll have to make it Linux-only. ;) Warnings don't generally bother me too much. I try things out on a test machine first anyway. If something labeled as experimental and dangerous passes my tests and works then I'll use it before something labeled as stable and released that I haven't tested. -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] When will Reiserfs be ready?
On Mon, 7 Jan 2002 12:42, _nasturtium wrote: All of the blathering and silliness removed. Short version: Russell made a few comments to explain some stuff on the assumption that Nasturtium was actually asking honest questions. Nasturtium made a number of ad hominem attacks. Blah blah blah. You have not actually addressed my reply, merely blather on. My main point was that someone on the MandrakeSoft QA team notes that you cannot write to NTFS5 partitions and that was what I posted. If someone, even one who has been on list for years, wants to doubt it I will reply with proof. That is wrong. From your original message on the topic of NTFS: DON'T swap to NTFS because Linux support is read-only. No reference is made to any particular version of NTFS or of Windows. No other qualifiying statement is made, you clearly and directly stated that NTFS was not writable on Linux. Stop trying to claim you were only referring to Windows XP, you said nothing of the kind in your original message. -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] Journal Questions
On Mon, 7 Jan 2002 13:11, pesarif wrote: 1. How big is the journal? 32M. It is possible to change this, but currently that requires recompiling your kernel (and running an altered mkreiserfs). Then a regular kernel won't mount them. It's painful enough that you don't want to do it. Hans has announced plans to address this issue, I am looking forward to a version of ReiserFS that works on floppies. ;) I have just made (today :)) a 600MB reiser 3.6 partition in Mandrake 8.0 and the usuable space reported by df -h, was 596MB. Isn't the journal 32MB because on my other system with a 5GB reiser 3.6 parition, 32MB is missing in df -h. And also, will the journal be larger with a larger filesystem? The journal is always the same size. As for the 600MB partition, are you certain that the partition was really 600M? Or did you just tell your fdisk program to make it 600M? Fdisk will always round up the sizes to the nearest cylinder and the difference can be 30M or more... 2. How do I disable journal replay and save/restore the journal or delete it completely? You can't. Ext3 has this functionality because it's an addition to Ext2 which doesn't have it. So you can switch between ext3 and ext2 by mounting it with a different driver. Also the journal is a file on ext3 so changing it's size is less invasive. We've all been hanging out for this for a long time. Unfortunately Hans has had other things to work on, no sponsor has demanded it, and no volunteer has come forward to do it. Maybe Hans will comment on where it is in the current schedule... PS It's an interesting co-incidence that we get two new users on the list from bigpond in the same week who both use Kmail 1.3... -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] magic is useless Determining File Types
On Fri, 4 Jan 2002 01:15, Alexander G. M. Smith wrote: Same thing for BeOS - floppies are FAT16 format (you can format for BFS but with the journal etc, there's 300KB of space for data), there's also FAT32 for Windows disk partitions and several other file systems. Some, like Mac HFS support a limited number of attributes (just the ones which have a Mac equivalent). Still, they got used by most of the regular applications written for BeOS, even if just used to specify the file type. Though if you used POSIX commands (like cp), the attributes would get lost. ZIP format If even cp doesn't support it then it's useless. This is why multiple streams were useless on NT because the cmd.exe copy command didn't support them (presumably nothing has changed with XP). So, if it's available and useful then there's a good chance people will use it in new software. When even the authors of the OS don't support it in their core file copy utility then it's not getting used much. On Fri, 4 Jan 2002 01:48, Jens Benecke wrote: Microsoft has these problems with their NTFS attributes. All the office type apps and so on were pressed hard to make heavy use of these attributes: you can e.g. view author, etc. of a MS-Word file in the file properties dialog, or the download URL of a .zip file, just like OS/2 did in 1996 :) but apart from that, nobody is really using these features, because you still *CAN* install Windows on FAT partitions and there you don't have these features. OS/2 had extended attributes in 1988. OS/2 had a fully object-oriented desktop using EAs in every imaginable way in 1992. By 1996 OS/2 was seriously losing market-share, mind-share, and IBM support. On Fri, 4 Jan 2002 09:48, Raphael Bosshard wrote: The idea of putting the filetype (ie. as MIME) into an additional file-attribute is not new and has done before by various systems, including OS/2, BeOS and even Windows. But in these cases, limitations of the FAT-Filesystem prevented an adoption of this feature. In the Unix-enviroment, it would fail because of standards and laziness; most of the file manipulating tools would have to be rewritten or to be patched. Right? Well, at least it was a nice idea... ;) I'm not sure it was such a nice idea really. Mainframes and mini-computers had typed files before Unix was invented. Unix was one of the earlier OSs to use strictly non-typed files (a file is just a collection of bytes). CP/M, DOS, etc all just followed that example. If we're going to experiment with new things, then how about indexed files managed by the file system which allow hardware devices such as EMC machines to do the database operations. This is why an IBM zSystem running OS/390 will beat almost anything for bulk IO while the same zSystem running Linux will apparently give poor IO performance. -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] When will Reiserfs be ready?
On Fri, 4 Jan 2002 11:14, _nasturtium wrote: I was reading the FAQ on www.namesys.com and it seems Reiser4 is sponsored (but not endorsed by...) by DARPA (Defence Advanced Research Projects Agency?). That seems like a good source of funds compared to your support business - your Support page only claims one request/payment every few days. You should try running a business when you grow up. Then you will discover the perilous position that having only a single customer can place you in. Does the name of Jeremy Fitzhardinge come to mind? While I recognise that most developers are paid, the aforementioned happens to be Volunteer. Author of hashing code. (teahash.c). Surely the open source model would allow more contributors. The model is open. The code is all released under the GPL and anyone who wishes can write new features or fix bugs. The reason why almost everyone who writes ReiserFS code works for Hans is that he appears to make a job offer to anyone who writes some ReiserFS code. Quite some time ago Chris Mason appeared from no-where, started contributing patches for ReiserFS, wrote the journalling code and got hired by Hans. There is always the option of forking ReiserFS if enough intelligent people believe that Hans is doing the wrong thing. So far there have been discussions about a number of issues, but the consensus of opinion among people who matter is that Hans is doing a reasonably good job. -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] funny file permission after reiserfsck
On Thu, 13 Dec 2001 04:15, W. Wilson Ho wrote: Hi all, After I've run reiserfsck on my disk, I have a file with 0 permission: # ls -l 0- 1 root root 238 Dec 11 22:52 lk This file is not readable. Adding rw permission to it does not make it readable again: # chmod a+rw lk # ls -l 0rw-rw-rw- 1 root root 238 Dec 11 22:52 lk # cat lk # From filemode.c as referenced by ls on my system, here's the list of prefix characters. No '0' character. /* Return a character indicating the type of file described by file mode BITS: 'd' for directories 'b' for block special files 'c' for character special files 'm' for multiplexor files 'M' for an off-line (regular) file 'l' for symbolic links 's' for sockets 'p' for fifos '-' for regular files '?' for any other file type. */ What distribution do you use? From my reading of the ls source I don't think it's possible to cause a leading '0' on Debian (of course I could have missed something, that source is painful to read). What happens when you run stat lk? -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
[reiserfs-list] per-char IO tests
I have released a new experimental version of Bonnie++ that includes a program to test per-char IO using putc()/getc(), putc()/getc() when linked in a non-threaded way (significantly improves performance) putc_unlocked()/getc_unlocked(), and write()/read(). Here's the results of testing my Thinkpad T20 with P3-650: Version 1.93 write read putcNT getcNT putc getc putcU getcU lyta 142651 8189 9348 1763 1813 22174 44887 lyta,142,651,8189,9348,1763,1813,22174,44887 Here's the results of testing my Athlon 800 play machine: Version 1.93 write read putcNT getcNT putc getc putcU getcU test 146607 7356 7280 1834 1971 41995 59100 test,146,607,7356,7280,1834,1971,41995,59100 Both machines run ReiserFS. A quick test indicates that using Ext2 instead of ReiserFS triples the performance of write(fd, buf, 1), but this is something I already knew (and had mentioned before on the ReiserFS list). Also based on previous tests I expect Solaris to outperform Linux with glibc on putcNT, getcNT, putc, and getc. The regular performance of putc() on Solaris comes close to putc_unlocked() on Linux with glibc. I'd like to thank Andrew Morton for forwarding messages from L-K that provoked me to write this new test program. I was tempted to subscribe to L-K to join this discussion, but it seems that Linus is saying everything that needs to be said anyway so there's no point. ;) -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] Linux 2.4.10 + reiserfs + raid
On Tue, 9 Oct 2001 07:32, Clement wrote: I just set up a box with RH 7.1, updated with Linux Kernel 2.4.10, compiled with reiserfs and raid support. I can define RAID-1 partitions and 'mkreiserfs /dev/md0' and mount/umount the partition with no trouble at all. However, once I reboot the machine, the raid partitions are not readable any more!!! What do you mean by not readable? Is ReiserFS unable to mount or is the RAID device not created? What does cat /proc/mdstat say? ReiserFS on RAID-1 on 2.4.10 is working perfectly for me (I have raidstart run from an initrd). -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] /etc/magic
On Mon, 16 Jul 2001 15:56, Nikita Danilov wrote: Russell Coker writes: Could someone please write up some /etc/magic entries for reiserfs? I think it should be something like: 0x10034 string ReIsErFs Old Reiserfs 0x10034 string ReIsEr2Fs New Reiserfs But that doesn't work for some reason... Take source distribution of file, increase HOWMANY in file.h (at least to 0x10050), recompile and it'll work. Stock file(1) only looks at the 16384 first bytes of the file. Thanks for that! By the way, are you considering sending your magic entry to the directory maintainer? If so, can you reword it to ReiserFS v3.5 and ReiserFS v3.6 otherwise we'll end with miserable Newest Reiserfs etc. Sure. I wrote that quickly after spending an hour fiddling with /etc/magic trying to get it to work! -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] /etc/magic
What do you think of the following? 0x10034 string ReIsErFs ReiserFS V3.5 0x10034 string ReIsEr2Fs ReiserFS V3.6 0x1002c short x block size %d 0x10032 short 2 (mounted or unclean) 0x1 long x num blocks %d 0x10040 long =1 tea hash 0x10040 long =2 yura hash 0x10040 long =3 r5 hash -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] CPU useage of ReiserFS
On Saturday 30 June 2001 20:29, Jens Benecke wrote: I just had a, er, 'lively' discussion with someone claiming ReiserFS is crap because it hogs even the fastest CPU too much, and it uses 4x as much processing power to do metadata operations, and in general is slower because of the journal. My benchmarks don't reflect this, especially on current hardware (ATA-66 and ATA-100 disks on VIA chipsets). While I agree that the journal does create an additional overhead, I'd like to know if the CPU overhead is really that much. I've seen your benchmarks on the web site but they don't say anything about CPU useage. I agree with Craig, I have one thing to add that Craig missed. Every 12 to 24 months CPU speed doubles. Now 1.4GHz CPUs with advanced cache and memory architectures are common while in 1990 20MHz CPUs without any caches were where it was at. Hard drive speed increases much more slowly. Now typical seek times are around 5ms and transfer rates are 35MB/s. In 1990 seek times were around 24ms and transfer rates were around 1MB/s. For future scalability a file system that uses lots of CPU time may be better than a file system that uses lots of disk access. -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] kernel-2.4.6-pre3 to 2.2.19 NFS tests
On Friday 22 June 2001 14:28, Christian Mayrhuber wrote: Try testing with Bonnie++, the file creation and deletion tests will give interesting results! ;) In one test I had an AIX machine that was moderately grunty (two fast POWER CPU's, 6 hard drives on a 160MB/s bus, 256M of RAM). When running Bonnie++ I found that the AIX machine could create files on my Thinkpad over a 10baseT NFS mount faster than it could create them on a local JFS file system! This is not the case for me, nfs performance never reaches local disk performance. Of course not. But ReiserFS on an NFS server can outperform local disks running other file systems. I think the network is the limiting factor. I don't have a idea if the bonnie file creation numbers over nfs are good ones or not, at least stat seems to be speedy. 1GHZ Athlon AMI Megaraid Raid-5 138GB/total, kernel-2.4.6-pre5, local, reiserfs Unknown,,30,14169,99,+,105,16224,91,13047,96,+,100, 14010,100 1GHz Athlon, client, Raid-5 array mounted over a 100MBit/s network Unknown,,30,3578,30,13676,53,4682,33,3614,31,17797,49,4251, 29 Now try a local Ext2 file system and see if you get more than 200 files created per second. Then try a local JFS file system on AIX and see if you can get more than 50 files created per second. ;) -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] kernel-2.4.6-pre3 to 2.2.19 NFS tests
On Thursday 14 June 2001 15:51, Christian Mayrhuber wrote: I've run bonnie on nfs over a 10MBit/s network on a ext2 and a reiserfs partition on the same disk. The Bad: The performance loss to ext2 on the same disk ist quit drastic, about 25% and this is only over a 10MBit/s network. What will happen on a 100MBit/s network? I have no chance to test it on 100MBit/s, SCSI hardware and a 3c59x card till monday. Try testing with Bonnie++, the file creation and deletion tests will give interesting results! ;) In one test I had an AIX machine that was moderately grunty (two fast POWER CPU's, 6 hard drives on a 160MB/s bus, 256M of RAM). When running Bonnie++ I found that the AIX machine could create files on my Thinkpad over a 10baseT NFS mount faster than it could create them on a local JFS file system! -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] optimizing reiserfs for large files?
On Thursday 14 June 2001 12:18, grobe wrote: I have a significant loss of performance in bonnie tests. The writing intelligently-test e.g. gives me 20710 kB/s with reiserfs, while I get 24753 kB/s with ext2 (1 GB-file). How much RAM do you have? If you have more than 512M of RAM then the results won't be a good indication of true performance. Also older versions of bonnie never sync the data so the performance report depends to a large extent on how much data remains in the write-back cache at the end of the test! Bonnie++ addresses these issue. Also neither of those results is what you should expect from modern hardware. Machines that were typically sold in corner stores about a year ago (such as the machine under my desk) return results better than that. I have attached the results of an Athlon-800 with 256M of PC-133 RAM and a single 46G ATA-66 IBM hard drive. The machine was not the most powerful machine on the market when I bought it over a year ago. What types of hard drives does the machine have? -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page Version 1.92b --Sequential Output-- --Sequential Input- --Random- Concurrency 1 -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- MachineSize K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP temp 496M 447 98 28609 16 10608 7 718 98 34694 15 199.8 1 Latency 22328us2074ms 56626us 57412us 43123us2984ms Version 1.92b --Sequential Create-- Random Create temp-Create-- --Read--- -Delete-- -Create-- --Read--- -Delete-- files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP 16 849 98 + +++ 15216 90 863 99 + +++ 3423 98 Latency 9168us 113us 249us 12778us 41us1744us 1.92b,1.92b,temp,1,993204157,496M,,447,98,28609,16,10608,7,718,98,34694,15,199.8,1,16,849,98,+,+++,15216,90,863,99,+,+++,3423,98,22328us,2074ms,56626us,57412us,43123us,2984ms,9168us,113us,249us,12778us,41us,1744us
Re: [reiserfs-list] reiserfs-raw
On Monday 18 June 2001 21:57, Henrik Nordstrom wrote: For Squid it would become very interesting if in some time (lets say about a year, maybe more) there is a good volatile permanent object store similar to reiserfs-raw but with a slightly more flexible application interface. One thing I have considered doing if I got a large amount of spare time (IE something that'll never happen) is to investigate getting the user-mode-linux code and taking the block IO part to make file systems run in user-space as a database interface. For something like a large squid box it might get a performance gain to have small operations (directory lookups) take place in user-land rather than have a system call for each one. Also it could potentially have some benefits for debugging. I thought that combining the above with a LD_PRELOAD library to take over the read/write/open/etc library calls could allow an application to think it's using regular files while it's really accessing a user-land process and talking over named pipes. Then you could test out a new version of a file system without risking crashing your machine! -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
Re: [reiserfs-list] optimizing reiserfs for large files?
On Saturday 23 June 2001 01:11, Lars O. Grobe wrote: Also neither of those results is what you should expect from modern hardware. Machines that were typically sold in corner stores about a year ago (such as the machine under my desk) return results better than that. I have attached the results of an Athlon-800 with 256M of PC-133 RAM and a single 46G ATA-66 IBM hard drive. The machine was not the most powerful machine on the market when I bought it over a year ago. What types of hard drives does the machine have? G should be quite fast sca-scsi ibm-drives. As I wrote, it's an 320GB array in a EXP15 connected to a IBM ServeRAID4M. The Netfinity has two 833MHz PIIIs. Hmm. Sounds like the performance you describe is less than expected, and the performance is being over-stated too! When you get some more accurate results it'll look even worse... -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page
[reiserfs-list] Released Bonnie++ 1.92a
This experimental version now has a -z command-line option to specify the seed for random number generation as suggested by several people on this list. When I do repeated runs of it I don't see results being any closer together than when using different random seeds. But it will hopefully be useful to you anyway. Also I have fixed a bug where the experimental versions would SEGV after completing the IO tests for no apparent reason. It is available on my web site (which I have just moved to a better connected server for faster transfers) and on Sourceforge. Another project that I am starting to work on is an LDAP benchmark. If you are interested in benchmarking LDAP servers or tuning ReiserFS for better OpenLDAP performance then you can talk to me off the list about what your requirements are (I am still in the design phase). -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page