[zfs-discuss] Re: ZFS vs. Apple XRaid
Please read also http://docs.info.apple.com/article.html?artnum=303503. This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Re: Veritas NetBackup Support for ZFS
I am using Netbackup 6.0 MP3 on several ZFS systems just fine. I think that NBU won't back up some exotic ACLs of ZFS, but if you are using ZFS like other filesystems (UFS, etc) then there aren't any issues. Hum. ACLs are not so exotic. This IS a really BIG issue. If you are using ACLs, even POSIX, moving production to ZFS filesystems means loosing any ACLs in backups. In other words, if you're using 30 years old UNIX rights, no problem. If I'd have to give a list of complaint on ZFS, that would be the first on my list ! Sun SHOULD make pressure on backup software editor (or send them some engineer) to support ZFS. This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Newbie in ZFS
Hi all, as I am newbie in ZFS, yesterday I played with it a little bit and there are so many good things but I've notes few things I couldn't explain so. 1) It's not possible anymore within a pool create a file system with a specific sizeIf I have 2 file systems I can't decide to give for example 10g to one and 20g to the other one unless I set a reservation for them. Also I tried to manually create pool with slices and have for each pool a FS with the size I wanted..Is that true? 2) I mirrored 2 disks within the same D1000 and while I was putting a big tar ball in the FS I tried to physically remove one mirror and ..I had to turn off the system as I couldn't login in even trough the console. Is there something wrong on what I did? cheers -- Alf ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Newbie in ZFS
Hi, Dick Davies wrote: On 22/09/06, Alf [EMAIL PROTECTED] wrote: 1) It's not possible anymore within a pool create a file system with a specific sizeIf I have 2 file systems I can't decide to give for example 10g to one and 20g to the other one unless I set a reservation for them. Also I tried to manually create pool with slices and have for each pool a FS with the size I wanted..Is that true? zfs set quota=5G poolname/fsname will give you a filesystem that shows up as 5GiB in 'df' - is that what you want? I tried quota and it works fine but if you have another fs that takes all the space you have in the pool, your FS will not have space for itself. So set a reservation for the FS is fine but comparing with other VOLUME MANAGER is differentthat's what I am asking! You mean pull it out? Does your hardware support hotswap? As far as I know D1000 support itdoes it? cheers Dick Davies wrote: On 22/09/06, Alf [EMAIL PROTECTED] wrote: 1) It's not possible anymore within a pool create a file system with a specific sizeIf I have 2 file systems I can't decide to give for example 10g to one and 20g to the other one unless I set a reservation for them. Also I tried to manually create pool with slices and have for each pool a FS with the size I wanted..Is that true? zfs set quota=5G poolname/fsname will give you a filesystem that shows up as 5GiB in 'df' - is that what you want? 2) I mirrored 2 disks within the same D1000 and while I was putting a big tar ball in the FS I tried to physically remove one mirror and You mean pull it out? Does your hardware support hotswap? -- Alfredo De Luca == May you live in interesting times. == ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Newbie in ZFS
Hi Michael, I completely agree with you. I was just wondering about the differences between ZFS and others VM and also if I got the essence of it. Also customers could ask these things and if they can use ZFS filesystems like old fashion mode setting a specific size. What do you thing about pulling out a mirror on D1000 and the completely hang of the system? -- Alf Michael Schuster wrote: Alf wrote: Hi, Dick Davies wrote: On 22/09/06, Alf [EMAIL PROTECTED] wrote: 1) It's not possible anymore within a pool create a file system with a specific sizeIf I have 2 file systems I can't decide to give for example 10g to one and 20g to the other one unless I set a reservation for them. Also I tried to manually create pool with slices and have for each pool a FS with the size I wanted..Is that true? zfs set quota=5G poolname/fsname will give you a filesystem that shows up as 5GiB in 'df' - is that what you want? I tried quota and it works fine but if you have another fs that takes all the space you have in the pool, your FS will not have space for itself. So set a reservation for the FS is fine but comparing with other VOLUME MANAGER is differentthat's what I am asking! with volume managers (ZFS is more than a VM), you don't get to share all the free space between file systems - that's one of the BIG advantages of zfs, IMO. I think using reservations is a very minor nuisance in comparison to the administrative effort you have to go to in other Volume Managers to move free space from one FS to another. -- Alfredo De Luca == May you live in interesting times. == ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Newbie in ZFS
Alf wrote: What do you thing about pulling out a mirror on D1000 and the completely hang of the system? I on purpose left that for others to answer - I don't know HW well enough by far :-) -- Michael Schuster +49 89 46008-2974 / x62974 visit the online support center: http://www.sun.com/osc/ Recursion, n.: see 'Recursion' ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Newbie in ZFS
Hi James, I agree. with you but I think it could take a while cheers Alf James C. McPherson wrote: Alf wrote: Hi Michael, I completely agree with you. I was just wondering about the differences between ZFS and others VM and also if I got the essence of it. Also customers could ask these things and if they can use ZFS filesystems like old fashion mode setting a specific size. That is part of the problem - ZFS _requires_ a complete re-working of your understanding of how storage works, because the old limitations are no longer valid. If the customer actually wants to get benefit from ZFS then they have to be prepared to undergo a paradigm shift. James C. McPherson -- Solaris kernel software engineer, system admin and troubleshooter http://www.jmcp.homeunix.com/blog Find me on LinkedIn @ http://www.linkedin.com/in/jamescmcpherson -- Alfredo De Luca == May you live in interesting times. == ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Newbie in ZFS
On 9/22/06, Dick Davies [EMAIL PROTECTED] wrote: On 22/09/06, Alf [EMAIL PROTECTED] wrote: 2) I mirrored 2 disks within the same D1000 and while I was putting a big tar ball in the FS I tried to physically remove one mirror and You mean pull it out? Does your hardware support hotswap? And even more to the point, do the Solaris drivers support hotswap on your hardware? When I was first inquring about hotswap hardware (about which I knew nothing then) nobody warned me about this, and I'm now the proud owner of a fine case with 8 hot-swap bays which work fine -- but it turns out that Solaris doesn't support hotswap on the SATA controllers on my motherboard, although it access the disks through them fine. In fact, it turns out (read recent postings) that even chipsets it claims to support are still being issued on new hardware in steppings that aren't actually supported. This area seems to be a major minefield currently. -- David Dyer-Bennet, mailto:[EMAIL PROTECTED], http://www.dd-b.net/dd-b/ RKBA: http://www.dd-b.net/carry/ Pics: http://www.dd-b.net/dd-b/SnapshotAlbum/ Dragaera/Steven Brust: http://dragaera.info/ ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Newbie in ZFS
You mean pull it out? Does your hardware support hotswap? As far as I know D1000 support itdoes it? I'm sure the D1000 is fine with the concept. It's probably something in the software stack that is upset. I was told that a similar issue that I once had when testing was likely due to some limitations in how ZFS and the sd driver communicates. That the sd driver will take a really long time to timeout each of what may be several I/Os to it. -- Darren Dunham [EMAIL PROTECTED] Senior Technical Consultant TAOShttp://www.taos.com/ Got some Dr Pepper? San Francisco, CA bay area This line left intentionally blank to confuse you. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: live upgrade incompability
I believe I am experiencing a similar, but more severe issue and I do not know how to resolve it. I used liveupgrade from s10u2 to NV b46 (via solaris express release). My second disk is zfs with the file system fitz. I did a 'zpool export fitz' Reboot with init 6 into new environment, NV b46, I get the following error: cannot mount '/fitz' : directory is not empty svc:/system/filesystem/local:default: WARNING: /usr/sbin/zfs mount -a failed: exit status 1 svc.startd[7]: svc:/system/filesystem/local:default: Method /lib/svc/method/fs-local failed with exit status 95. zfs list = nothing listed. There is already a /fitz directory filled with the zpool fitz files on mounted. Since filesystem/local svc won't start, I cannot start X, which is critical to using the computer. I now see that there was no real need to export the pool fitz and that I should have just imported it once in the new BE. How can I now solve this issue? (BTW, attempting to boot back into s10u2, the original BE, results in a kernel panic, so I cannot go back). thanks, aric --On September 21, 2006 10:01:28 AM -0700 Haik Aftandilian [EMAIL PROTECTED] sent: I did a liveupgrade from NV b41 to b47 and I still ran into this problem on one of my ZFS mounts. Both mounts failed to mount in the new BE because directories were created for the mount points, but only one of the mounts actually had its data copied into the BE. I checked /etc/default/lu and I do have the fix for 6335531 Liveupgrade should not copy zfs file systems into new BEs which was putback to build 27. Here's my configuration # zfs list NAME USED AVAIL REFER MOUNTPOINT scratch 3.07G 39.5G 3.07G /scratch twosyncs 52.1G 176G 24.5K /twosyncs twosyncs/home 52.1G 176G 25.5K /export/home twosyncs/home/haik52.1G 176G 51.4G /export/home/haik The data in /scratch was copied into a /scratch directory in the new BE. /export/home/haik wasn't copied into the new BE, but directories were created in the new BE preventing it from mounting on boot. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Re: Re: live upgrade incompability
I believe I am experiencing a similar, but more severe issue and I do not know how to resolve it. I used liveupgrade from s10u2 to NV b46 (via solaris express release). My second disk is zfs with the file system fitz. I did a 'zpool export fitz' Reboot with init 6 into new environment, NV b46, I get the following error: cannot mount '/fitz' : directory is not empty svc:/system/filesystem/local:default: WARNING: /usr/sbin/zfs mount -a failed: exit status 1 svc.startd[7]: svc:/system/filesystem/local:default: Method /lib/svc/method/fs-local failed with exit status 95. zfs list = nothing listed. There is already a /fitz directory filled with the zpool fitz files on mounted. Since filesystem/local svc won't start, I cannot start X, which is critical to using the computer. I now see that there was no real need to export the pool fitz and that I should have just imported it once in the new BE. How can I now solve this issue? (BTW, attempting to boot back into s10u2, the original BE, results in a kernel panic, so I cannot go back). Aric, It sounds like you can resolve this issue by simply booting into the new BE and deleting the /fitz directory and then rebooting and going back into the new BE. I say this because from your message it sounds like the data from your zfs filesystem in /fitz was copied to /fitz in the new BE (instead of just being mounted in the new BE). BEFORE DELETING ANYTHING, please make sure /fitz is not a zfs mount and just a plain directory and therefore just a copy of what is in your zpool. Be careful, I don't want you to lose any data. Also, what does zpool list report? Lastly, ZFS people might be interested in the panic message you get when you boot back into Solaris 10. Haik thanks, aric This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: live upgrade incompability
Apologies for any confusion, but I am now able to give more output regarding the zpool fitz. unknown# zfs list -- returns list of zfs file system fitz and related snapshots unknown# zpool status pool: fitz state: ONLINE status: The pool is formatted using an older on-disk format. The pool can still be used, but some features are unavailable. action: Upgrade the pool using 'zpool upgrade'. Once this is done, the pool will no longer be accessible on older software versions. scrub: none requested config: NAMESATE fitzONLINE c2d0s7 ONLINE errors: No known data errors unknown# zpool upgrade -v This system is currently running ZFS version 3. the following versions are supported: .. unknown# zfs mount -- lists the zfs pool as mounted as it should be at /fitz but 'zfs unmount fitz' returns 'cannot unmount 'fitz' : not currently mounted zpool import -- no pools available to import zpool import -d /fitz -- no pools available to import thanks, aric I believe I am experiencing a similar, but more severe issue and I do not know how to resolve it. I used liveupgrade from s10u2 to NV b46 (via solaris express release). My second disk is zfs with the file system fitz. I did a 'zpool export fitz' Reboot with init 6 into new environment, NV b46, I get the following error: cannot mount '/fitz' : directory is not empty svc:/system/filesystem/local:default: WARNING: /usr/sbin/zfs mount -a failed: exit status 1 svc.startd[7]: svc:/system/filesystem/local:default: Method /lib/svc/method/fs-local failed with exit status 95. zfs list = nothing listed. There is already a /fitz directory filled with the zpool fitz files on mounted. Since filesystem/local svc won't start, I cannot start X, which is critical to using the computer. I now see that there was no real need to export the pool fitz and that I should have just imported it once in the new BE. How can I now solve this issue? (BTW, attempting to boot back into s10u2, the original BE, results in a kernel panic, so I cannot go back). thanks, aric ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Re: Building large home file server with SATA
Alexei Rodriguez wrote: Unless they break the spec, yes, it should work. PCI Excellent to know! I will verify that the motherboard and the PCI-X cards play well together. Thanks! Alexei This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: Building large home file server with SATA
On September 22, 2006 10:26:01 AM -0700 Alexei Rodriguez [EMAIL PROTECTED] wrote: Alexei Rodriguez wrote: Unless they break the spec, yes, it should work. PCI Excellent to know! I will verify that the motherboard and the PCI-X cards play well together. You might run into a problem with 3.3V cards vs 5V slots. -frank ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] slow reads question...
I have set up a small box to work with zfs. (2x 2.4GHz xeons, 4GB memory, 6x scsi disks) I made one drive the boot drive and put the other five into a pool with the zpool create tank command right out of the admin manual. The administration experience has been very nice and most everything as worked as expected. (Setting up new filesystems, swapping out failed drives, etc.) What isnt as I expected is the slow speed. When using a raw device, a scsi disk on the system reads at 34MB/s. About what I would expect for these disks. | # time dd if=/dev/rdsk/c0t1d0 of=/dev/null bs=8k count=102400 | 102400+0 records in | 102400+0 records out | | real0m23.182s | user0m0.135s | sys 0m1.979s However when reading from a 10GB file of zeros, made with mkfile, the read performace is much lower, 11MB/s. | # time dd if=zeros-10g of=/dev/null bs=8k count=102400 | 102400+0 records in | 102400+0 records out | | real1m8.763s | user0m0.104s | sys 0m1.759s After reading the list archives, I saw ztune.sh. Using it I tried a couple of different settings and didnt see any changes. After that I toggled the compression, atime, recordsize and checksum options on and off to no avail. Am I expecting too much from this setup? What might be changed to speed things up? Wait until snv_45? The version of open solaris is: | # uname -a | SunOS donatella 5.11 snv_44 i86pc i386 i86pc The options on the filesystem are: | # zfs get all tank/home | NAME PROPERTY VALUE SOURCE | tank/home type filesystem - | tank/home creation Fri Sep 22 10:47 2006 - | tank/home used 39.1K - | tank/home available 112G - | tank/home referenced 39.1K - | tank/home compressratio 1.00x - | tank/home mountedyes- | tank/home quota none default | tank/home reservationnone default | tank/home recordsize 128K default | tank/home mountpoint /export/zfslocal | tank/home sharenfs on local | tank/home checksum on default | tank/home compressionoffdefault | tank/home atime on default | tank/home deviceson default | tank/home exec on default | tank/home setuid on default | tank/home readonly offdefault | tank/home zoned offdefault | tank/home snapdirhidden default | tank/home aclmodegroupmask default | tank/home aclinherit secure default thanks, harley. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Re: slow reads question...
ZFS uses a 128k block size. If you change dd to use a bs=128k, do you observe any performance improvement? | # time dd if=zeros-10g of=/dev/null bs=8k count=102400 | 102400+0 records in | 102400+0 records out | real1m8.763s | user0m0.104s | sys 0m1.759s It's also worth noting that this dd used less system and user time than the read from the raw device, yet took a longer time in real time. This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] I'm dancin' in the streets
Wow! I solved a tricky problem this morning thanks to Zones ZFS integration. We have a SAS SPDS database environment running on Sol10 06/06. The SPDS database is unique in that when a table is being updated by one user it is unavailable to the rest of the user community. Our nightly update jobs (occassionally they turn into day jobs when they take longer :-() were coming in the way of our normal usage. So I put on my ZFS cap and figure it can be simply solved by deploying the 'clone' feature. Simply stated I'd create a clone of all the SPDS filesystems and start another instance of SPDS to read/write from the cloned data. Unfortunately I hit a wall when I realized that there is no way to update the SPDS metadata (binary file containing a description of the physical structure of the database) with the new directory path. I was stumped until it occurred to me that I can solve it by simply marrying the clones with a Solaris Zone Now our problem is solved as follows: 1. Stop local zone 2. Reclaim the ZFS clones in the global-zone 3. Destroy the clone/snapshot 4. Recreate the clone/snapshot 5. Restart the local zone 6. Start SPDS in the local zone and it works beautifully because it sees all the files it needs per its metadata!!! To accomplish the same in traditional methods would have required a SAN disk, disk merge/split, ... You get the picture, ugly! Chalk one more victory for the Solaris 10 Zones/ZFS!!! Thanks to the developers of these features that enabled me elegantly solve a difficult problem. -Anantha- This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] zpool always thinks it's mounted on another system
The history is quite simple: 1) Installed nv_b32 or around there on a zeroed drive. Created this ZFS pool for the first time. 2) Non-live upgraded to nv_b42 when it came out, zpool upgrade on the zpool in question from v2 to v3. 3) Tried to non-live upgrade to nv_b44, upgrade failed every time, so I just blew away my existing partition scheme and install nv_b44 cleanly. 4) Problem begins. I can't think of any sane reason I could have blown away that directory accidentally, so I don't know. - Rich On 9/22/06, Eric Schrock [EMAIL PROTECTED] wrote: On Fri, Sep 22, 2006 at 03:36:36AM -0400, Rich wrote: ...huh. So /etc/zfs doesn't exist. At all. Creating /etc/zfs using mkdir, then importing the pool with zpool import -f, then rebooting, the behavior vanishes, so...yay. Problem solved, I guess, but shouldn't ZFS be smarter about creating its own config directory? That seems a reasonable RFE, but I wonder how you got into this situation in the first place. What is the history of the OS on this system? Nevada? Solaris 10? Upgraded? Patched? I assume that you don't tend to go around removing random /etc directories on purpose, so I want to make sure that our software didn't screw up somehow. - Eric -- Eric Schrock, Solaris Kernel Development http://blogs.sun.com/eschrock -- Friends don't let friends use Internet Explorer or Outlook. Choose something better. www.mozilla.org www.getfirefox.com www.getthunderbird.com ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: slow reads question...
On Fri, 22 Sep 2006, johansen wrote: ZFS uses a 128k block size. If you change dd to use a bs=128k, do you observe any performance improvement? I had tried other sizes with much the same results, but hadnt gone as large as 128K. With bs=128K, it gets worse: | # time dd if=zeros-10g of=/dev/null bs=128k count=102400 | 81920+0 records in | 81920+0 records out | | real2m19.023s | user0m0.105s | sys 0m8.514s It's also worth noting that this dd used less system and user time than the read from the raw device, yet took a longer time in real time. I think some of the blocks might be cached, as I have run this a number of times. I really dont know how the time might be accounted for -- However, the real time is correct as that is what I see while waiting for the command to complete. Is there any other info I can provide which would help? harley. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Re: Re: live upgrade incompability
Haik, Thank you very much. 'zpool list' yeilds NAMESIZEUSEDAVAILCAPHEALTH ALTROOT z 74.5G 22.9G 51.6G30%ONLINE - How do I confirm that /fitz is not currently a zfs mountpoint? 'zfs mount' yeilds fitz/home/fitz/home fitz/home/aorchid /fitz/home/aorchid fitz/music /fitz/music fitz/pg/fitz/pg fitz/pictures/fitz/pictures Ah, OK. It's good that you didn't delete /fitz. This is what I recommend that you do. # zfs unmount -a # zfs mount This should produce no output since now all zfs filesystems are unmounted # find /fitz This should produce no files, only empty directories At this point, as long as there is nothing important in /fitz, you can go ahead an delete it # rm -r /fitz Or just delete everything inside /fitz # zfs mount -a Now /fitz should be all set. When you reboot you should not see the /fitz filesystem mount error Someone else please chime in if this looks wrong. Hope that helps. Haik 'ls -la /fitz' yeilds total 85 drwxr-xr-x 7 root sys 512 Sep 20 10:41 . drwxr-xr-x 30 root root 512 Sep 21 18:28 .. -- this is when I ran 'zpool export fitz' drwxr-xr-x 3 root sys 3 Jul 25 12:22 home etc... /etc/vfstab does not have /fitz and umount /fitz returns umount: warning: /fitz not in mnttab umount: /fitz not mounted Lastly, ZFS people might be interested in the panic message you get when you boot back into Solaris 10. They are all related to the NVIDIA driver, gfxp, from what I remember from two weeks ago. I am on an Ultra 20. thanks, aric This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: slow reads question...
Harley: I had tried other sizes with much the same results, but hadnt gone as large as 128K. With bs=128K, it gets worse: | # time dd if=zeros-10g of=/dev/null bs=128k count=102400 | 81920+0 records in | 81920+0 records out | | real2m19.023s | user0m0.105s | sys 0m8.514s I may have done my math wrong, but if we assume that the real time is the actual amount of time we spent performing the I/O (which may be incorrect) haven't you done better here? In this case you pushed 81920 128k records in ~139 seconds -- approx 75437 k/sec. Using ZFS with 8k bs, you pushed 102400 8k records in ~68 seconds -- approx 12047 k/sec. Using the raw device you pushed 102400 8k records in ~23 seconds -- approx 35617 k/sec. I may have missed something here, but isn't this newest number the highest performance so far? What does iostat(1M) say about your disk read performance? Is there any other info I can provide which would help? Are you just trying to measure ZFS's read performance here? It might be interesting to change your outfile (of) argument and see if we're actually running into some other performance problem. If you change of=/tmp/zeros does performance improve or degrade? Likewise, if you write the file out to another disk (UFS, ZFS, whatever), does this improve performance? -j ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Re: Re: low disk performance
Update ... iostat output during zpool scrub extended device statistics device r/sw/s Mr/s Mw/s wait actv svc_t %w %b sd34 2.0 395.20.10.6 0.0 34.8 87.7 0 100 sd3521.0 312.21.22.9 0.0 26.0 78.0 0 79 sd3620.01.01.20.0 0.0 0.7 31.4 0 13 sd3720.01.01.00.0 0.0 0.7 35.1 0 21 sd34 is always at 100% ... This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Re: Re: low disk performance
Update ... iostat output during zpool scrub extended device statistics w/s Mr/s Mw/s wait actv svc_t %w %b 34 2.0 395.20.10.6 0.0 34.8 87.7 0 100 3521.0 312.21.22.9 0.0 26.0 78.0 0 79 3620.01.01.20.0 0.0 0.7 31.4 0 13 3720.01.01.00.0 0.0 0.7 35.1 0 21 sd34 is always at 100% ... pool: zpool1 state: ONLINE scrub: scrub in progress, 0.13% done, 72h39m to go config: NAME STATE READ WRITE CKSUM zpool1ONLINE 0 0 0 raidzONLINE 0 0 0 c4t60001FE100118DB91190724700C7d0 ONLINE 0 0 0 c4t60001FE100118DB91190724700C9d0 ONLINE 0 0 0 c4t60001FE100118DB91190724700CBd0 ONLINE 0 0 0 c4t60001FE100118DB91190724700CCd0 ONLINE 0 0 0 72hours?? isn't too much for 370GB of data? This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: slow reads question...
On Fri, 22 Sep 2006, [EMAIL PROTECTED] wrote: Are you just trying to measure ZFS's read performance here? That is what I started looking at. We scrounged around and found a set of 300GB drives to replace the old ones we started with. Comparing these new drives to the old ones: Old 36GB drives: | # time mkfile -v 1g zeros-1g | zeros-1g 1073741824 bytes | | real2m31.991s | user0m0.007s | sys 0m0.923s Newer 300GB drives: | # time mkfile -v 1g zeros-1g | zeros-1g 1073741824 bytes | | real0m8.425s | user0m0.010s | sys 0m1.809s At this point I am pretty happy. I am wondering if there is something other than capacity and seek time which has changed between the drives. Would a different scsi command set or features have this dramatic a difference? thanks!, harley. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: slow reads question...
Harley: Old 36GB drives: | # time mkfile -v 1g zeros-1g | zeros-1g 1073741824 bytes | | real2m31.991s | user0m0.007s | sys 0m0.923s Newer 300GB drives: | # time mkfile -v 1g zeros-1g | zeros-1g 1073741824 bytes | | real0m8.425s | user0m0.010s | sys 0m1.809s This is a pretty dramatic difference. What type of drives were your old 36g drives? I am wondering if there is something other than capacity and seek time which has changed between the drives. Would a different scsi command set or features have this dramatic a difference? I'm hardly the authority on hardware, but there are a couple of possibilties. Your newer drives may have a write cache. It's also quite likely that the newer drives have a faster speed of rotation and seek time. If you subtract the usr + sys time from the real time in these measurements, I suspect the result is the amount of time you were actually waiting for the I/O to finish. In the first case, you spent 99% of your total time waiting for stuff to happen, whereas in the second case it was only ~86% of your overall time. -j ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: Re: low disk performance
On 9/22/06, Gino Ruopolo [EMAIL PROTECTED] wrote: Update ... iostat output during zpool scrub extended device statistics w/s Mr/s Mw/s wait actv svc_t %w %b 34 2.0 395.20.10.6 0.0 34.8 87.7 0 100 3521.0 312.21.22.9 0.0 26.0 78.0 0 79 3620.01.01.20.0 0.0 0.7 31.4 0 13 3720.01.01.00.0 0.0 0.7 35.1 0 21 sd34 is always at 100% ... pool: zpool1 state: ONLINE scrub: scrub in progress, 0.13% done, 72h39m to go config: NAME STATE READ WRITE CKSUM zpool1ONLINE 0 0 0 raidzONLINE 0 0 0 c4t60001FE100118DB91190724700C7d0 ONLINE 0 0 0 c4t60001FE100118DB91190724700C9d0 ONLINE 0 0 0 c4t60001FE100118DB91190724700CBd0 ONLINE 0 0 0 c4t60001FE100118DB91190724700CCd0 ONLINE 0 0 0 72hours?? isn't too much for 370GB of data? This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss For what it's worth, I've found that usually, within the first ~5m or so of starting a scrub, the time estimate is disproportionate to the actual time the scrub will take. - Rich ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss