[zfs-discuss] ZFS - SWAP and lucreate..
Good morning everybody I was migrating my ufs – rootfilesystem to a zfs – one, but was a little upset finding out that it became bigger (what was clear because of the swap and dump size). Now I am questioning myself if it is possible to set the swap and dump size by using the lucreate – command (I wanna try it again but on less space). Unfortunately I did not find any advice in manpages. Maybe you can help me? -- 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 import issue
Hi, you have to upgrade your pool: The pool is formatted using an older on-disk version. # *zpool upgrade -v* Then it should works fine. Kind regards, Moutacim Ketan schrieb: I'm having following issue .. i import the zpool and it shows pool imported correctly but after few seconds when i issue command zpool list .. it does not show any pool and when again i try to import it says device is missing in pool .. what could be the reason for this .. and yes this all started after i upgraded the powerpath abc # zpool import pool: emcpool1 id: 5596268873059055768 state: ONLINE status: The pool is formatted using an older on-disk version. action: The pool can be imported using its name or numeric identifier, though some features will not be available without an explicit 'zpool upgrade'. config: emcpool1 ONLINE emcpower0c ONLINE abc # zpool list no pools available -- Moutacim LACHHAB Service Engineer Software Technical Services Center Sun Microsystems Inc. Email moutacim.lach...@sun.com mailto:moutacim.lach...@sun.com +33(0)134030594 x31457 For knowledge and support: http://sunsolve.sun.com ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] zpool import issue
On 29.06.09 11:41, Ketan wrote: I'm having following issue .. i import the zpool and it shows pool imported correctly 'zpool import' only show what pools are available to import. In order to actually import pool you need to to zpool import emcpool1 but after few seconds when i issue command zpool list .. it does not show any pool this is expected as you did not do import yet. and when again i try to import it says device is missing in pool .. what could be the reason for this .. and yes this all started after i upgraded the powerpath I suspect PowerPath may be playing tricks with you, but cannot comment any further because there's no more data to comment on... This is not related to pool on-disk version though. victor abc # zpool import pool: emcpool1 id: 5596268873059055768 state: ONLINE status: The pool is formatted using an older on-disk version. action: The pool can be imported using its name or numeric identifier, though some features will not be available without an explicit 'zpool upgrade'. config: emcpool1 ONLINE emcpower0c ONLINE abc # zpool list no pools available ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] zpool import issue
didn't help .. tried r...@essapl020-u006 # zpool import pool: emcpool1 id: 5596268873059055768 state: UNAVAIL status: One or more devices are missing from the system. action: The pool cannot be imported. Attach the missing devices and try again. see: http://www.sun.com/msg/ZFS-8000-3C config: emcpool1 UNAVAIL insufficient replicas emcpower0c UNAVAIL cannot open r...@essapl020-u006 # zpool upgrade -v This system is currently running ZFS pool version 10. The following versions are supported: VER DESCRIPTION --- 1 Initial ZFS version 2 Ditto blocks (replicated metadata) 3 Hot spares and double parity RAID-Z 4 zpool history 5 Compression using the gzip algorithm 6 bootfs pool property 7 Separate intent log devices 8 Delegated administration 9 refquota and refreservation properties 10 Cache devices For more information on a particular version, including supported releases, see: http://www.opensolaris.org/os/community/zfs/version/N Where 'N' is the version number. r...@essapl020-u006 # zpool upgrade emcpool1 This system is currently running ZFS pool version 10. cannot open 'emcpool1': no such pool r...@essapl020-u006 # zpool upgrade This system is currently running ZFS pool version 10. All pools are formatted using this version. r...@essapl020-u006 # -- 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 import issue
And i just found out that one of my disk in the pool is showing missing labels r...@essapl020-u006 # zdb -l /dev/dsk/emcpower0c LABEL 0 version=4 name='emcpool1' state=0 txg=6973090 pool_guid=5596268873059055768 hostid=2228473662 hostname='essapl020-u006' top_guid=3858675847091731383 guid=3858675847091731383 vdev_tree type='disk' id=0 guid=3858675847091731383 path='/dev/dsk/emcpower0c' phys_path='/pseudo/e...@0:c,blk' whole_disk=0 metaslab_array=14 metaslab_shift=32 ashift=9 asize=477788372992 is_log=0 LABEL 1 version=4 name='emcpool1' state=0 txg=6973090 pool_guid=5596268873059055768 hostid=2228473662 hostname='essapl020-u006' top_guid=3858675847091731383 guid=3858675847091731383 vdev_tree type='disk' id=0 guid=3858675847091731383 path='/dev/dsk/emcpower0c' phys_path='/pseudo/e...@0:c,blk' whole_disk=0 metaslab_array=14 metaslab_shift=32 ashift=9 asize=477788372992 is_log=0 LABEL 2 failed to read label 2 LABEL 3 is there anyway to recover it ? -- This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] ZFS and Dinamic Stripe
Hello. I have a question about how ZFS works with Dinamic Stripe. Well, start with the next situation: - 4 Disk of 100MB in stripe format under ZFS. - We use the stripe in a 75%, so we have free 100MB. (easy) Well, we add a new disk of 100MB in the pool. So we have 200MB free but only 100MB will have the speed of 4 disk and, the rest 100MB will have the speed of 1 disk. The questions are: - Have ZFS any kind of reorganization of the data in the stripe that change this situation and become in 200MB free with the speed of 5 disks? - If the answer is yes, how is it does? in the background? Thanks for your time (and sorry for my english). JLBG -- This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] ZFS disk label missing
I upgraded powepath on my system and exported the zfs pool after upgradation i was able to import the pool but after reboot i 'm not able to import the pool and it fails with error zpool import emcpool1 cannot import 'emcpool1': invalid vdev configuration and digging a lil bit into it i found following things Jun 29 2009 01:54:14.928954823 ereport.fs.zfs.vdev.open_failed nvlist version: 0 class = ereport.fs.zfs.vdev.open_failed ena = 0xe503fd84f321 detector = (embedded nvlist) nvlist version: 0 version = 0x0 scheme = zfs pool = 0x4da9f0e3c642b898 vdev = 0x358cc53d21483bb7 (end detector) pool = emcpool1 pool_guid = 0x4da9f0e3c642b898 pool_context = 1 pool_failmode = wait vdev_guid = 0x358cc53d21483bb7 vdev_type = disk vdev_path = /dev/dsk/emcpower0c parent_guid = 0x4da9f0e3c642b898 parent_type = root prev_state = 0x1 __ttl = 0x1 __tod = 0x4a486516 0x375eb9c7 and zdb for this vdev i get r...@essapl020-u006 # zdb -l /dev/dsk/emcpower0c LABEL 0 version=4 name='emcpool1' state=0 txg=6973090 pool_guid=5596268873059055768 hostid=2228473662 hostname='essapl020-u006' top_guid=3858675847091731383 guid=3858675847091731383 vdev_tree type='disk' id=0 guid=3858675847091731383 path='/dev/dsk/emcpower0c' phys_path='/pseudo/e...@0:c,blk' whole_disk=0 metaslab_array=14 metaslab_shift=32 ashift=9 asize=477788372992 is_log=0 LABEL 1 version=4 name='emcpool1' state=0 txg=6973090 pool_guid=5596268873059055768 hostid=2228473662 hostname='essapl020-u006' top_guid=3858675847091731383 guid=3858675847091731383 vdev_tree type='disk' id=0 guid=3858675847091731383 path='/dev/dsk/emcpower0c' phys_path='/pseudo/e...@0:c,blk' whole_disk=0 metaslab_array=14 metaslab_shift=32 ashift=9 asize=477788372992 is_log=0 LABEL 2 failed to read label 2 LABEL 3 failed to read label 3 r...@essapl020-u006 # si there any way i can recover it without loosing my data on it ? ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS and Dinamic Stripe
Jose Luis Barquín Guerola wrote: Hello. I have a question about how ZFS works with Dinamic Stripe. Well, start with the next situation: - 4 Disk of 100MB in stripe format under ZFS. - We use the stripe in a 75%, so we have free 100MB. (easy) Well, we add a new disk of 100MB in the pool. So we have 200MB free but only 100MB will have the speed of 4 disk and, the rest 100MB will have the speed of 1 disk. The questions are: - Have ZFS any kind of reorganization of the data in the stripe that change this situation and become in 200MB free with the speed of 5 disks? - If the answer is yes, how is it does? in the background? Thanks for your time (and sorry for my english). JLBG When you add more vdevs to the zpool, NEW data is written to the new stripe width. That is, when data was written to the original pool, it was written across 4 drives. It now will be written across 5 drives. Existing data WILL NOT be changed. So, for a zpool 75% full, you will NOT get to immediately use the first 75% of the new vdevs added. Thus, in your case, you started with a 400MB zpool (with 300MB of data). You added another 100MB vdev, resulting in a 500MB zpool. 300MB is written across 4 drives, and will have the appropriate speed. 75% of the new vdev isn't immediately usable (as it corresponds to the 75% in-use on the other 4 vdevs), so you effectively only have added 25MB of immediately usable space. Thus, you have: 300MB across 4 vdevs 125MB across 5 vdevs 75MB wasted space on 1 vdev To correct this - that is, to recover the 75MB of wasted space and to move the 300MB from spanning 4 vdevs to spanning 5 vdevs - you need to re-write the entire existing data space. Right now, there is no background or other automatic method to do this. 'cp -rp' or 'rsync' is a good idea. We really should have something like 'zpool scrub' do this automatically. -- Erik Trimble Java System Support Mailstop: usca22-123 Phone: x17195 Santa Clara, CA ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS - SWAP and lucreate..
# *swap -d /dev/zvol/dsk/rpool/swap* # *zfs volsize=8G rpool/swap* # *swap -a /dev/zvol/dsk/rpool/swap* I'm still a bit fuzzy about how swap/dump and ZFS. If I have a pool: pool: pool1 state: ONLINE scrub: none requested config: NAMESTATE READ WRITE CKSUM pool1 ONLINE 0 0 0 c3t0d0s1 ONLINE 0 0 0 which occupies nearly the entire disk, should I allocate both swap and dump in the same pool? Right now I have a swap on a second disk: # swap -l swapfile dev swaplo blocks free /dev/dsk/c3t1d0s0 32,8 16 8395184 8395184 Would I do something like: # zfs volsize=4G pool1/swap # swap -a /dev/zvol/dsk/pool1/swap # swap -d /dev/dsk/c3t1d0s0 What about a dump, any recommendations? Thanks... -- 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] ZFS disk label missing
This can occur if the location of the end of the partition has changed locations. This could be due to the partition actually shrinking or if more than one partition references the same starting block, but different ending locations. Check your partition configuration in format and debug with zdb -l When you can see all 4 labels, then you should be able to import. -- richard Vibhor Neb wrote: I upgraded powepath on my system and exported the zfs pool after upgradation i was able to import the pool but after reboot i 'm not able to import the pool and it fails with error zpool import emcpool1 cannot import 'emcpool1': invalid vdev configuration and digging a lil bit into it i found following things Jun 29 2009 01:54:14.928954823 ereport.fs.zfs.vdev.open_failed nvlist version: 0 class = ereport.fs.zfs.vdev.open_failed ena = 0xe503fd84f321 detector = (embedded nvlist) nvlist version: 0 version = 0x0 scheme = zfs pool = 0x4da9f0e3c642b898 vdev = 0x358cc53d21483bb7 (end detector) pool = emcpool1 pool_guid = 0x4da9f0e3c642b898 pool_context = 1 pool_failmode = wait vdev_guid = 0x358cc53d21483bb7 vdev_type = disk vdev_path = /dev/dsk/emcpower0c parent_guid = 0x4da9f0e3c642b898 parent_type = root prev_state = 0x1 __ttl = 0x1 __tod = 0x4a486516 0x375eb9c7 and zdb for this vdev i get r...@essapl020-u006 # zdb -l /dev/dsk/emcpower0c LABEL 0 version=4 name='emcpool1' state=0 txg=6973090 pool_guid=5596268873059055768 hostid=2228473662 hostname='essapl020-u006' top_guid=3858675847091731383 guid=3858675847091731383 vdev_tree type='disk' id=0 guid=3858675847091731383 path='/dev/dsk/emcpower0c' phys_path='/pseudo/e...@0:c,blk' whole_disk=0 metaslab_array=14 metaslab_shift=32 ashift=9 asize=477788372992 is_log=0 LABEL 1 version=4 name='emcpool1' state=0 txg=6973090 pool_guid=5596268873059055768 hostid=2228473662 hostname='essapl020-u006' top_guid=3858675847091731383 guid=3858675847091731383 vdev_tree type='disk' id=0 guid=3858675847091731383 path='/dev/dsk/emcpower0c' phys_path='/pseudo/e...@0:c,blk' whole_disk=0 metaslab_array=14 metaslab_shift=32 ashift=9 asize=477788372992 is_log=0 LABEL 2 failed to read label 2 LABEL 3 failed to read label 3 r...@essapl020-u006 # si there any way i can recover it without loosing my data on it ? ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS and Dinamic Stripe
Erik Trimble wrote: Jose Luis Barquín Guerola wrote: Hello. I have a question about how ZFS works with Dinamic Stripe. Well, start with the next situation: - 4 Disk of 100MB in stripe format under ZFS. - We use the stripe in a 75%, so we have free 100MB. (easy) Well, we add a new disk of 100MB in the pool. So we have 200MB free but only 100MB will have the speed of 4 disk and, the rest 100MB will have the speed of 1 disk. The questions are: - Have ZFS any kind of reorganization of the data in the stripe that change this situation and become in 200MB free with the speed of 5 disks? - If the answer is yes, how is it does? in the background? Yes, new writes are biased towards the more-empty vdev. Thanks for your time (and sorry for my english). JLBG When you add more vdevs to the zpool, NEW data is written to the new stripe width. That is, when data was written to the original pool, it was written across 4 drives. It now will be written across 5 drives. Existing data WILL NOT be changed. So, for a zpool 75% full, you will NOT get to immediately use the first 75% of the new vdevs added. Thus, in your case, you started with a 400MB zpool (with 300MB of data). You added another 100MB vdev, resulting in a 500MB zpool. 300MB is written across 4 drives, and will have the appropriate speed. 75% of the new vdev isn't immediately usable (as it corresponds to the 75% in-use on the other 4 vdevs), so you effectively only have added 25MB of immediately usable space. Thus, you have: 300MB across 4 vdevs 125MB across 5 vdevs 75MB wasted space on 1 vdev To correct this - that is, to recover the 75MB of wasted space and to move the 300MB from spanning 4 vdevs to spanning 5 vdevs - you need to re-write the entire existing data space. Right now, there is no background or other automatic method to do this. 'cp -rp' or 'rsync' is a good idea. We really should have something like 'zpool scrub' do this automatically. No. Dynamic striping is not RAID-0, which is what you are describing. In a dynamic stripe, the data written is not divided up amongst the current devices in the stripe. Rather, data is chunked and written to the vdevs. When about 500 kBytes has been written to a vdev, the next chunk is written to another vdev. The choice of which vdev to go to next is based, in part, on the amount of free space available on the vdev. So you get your cake (stochastic spreading of data across vdevs) and you get to eat it (use all available space), too. -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Best controller card for 2 to 4 SATA drives ?
I've had success with the SIIG SC-SAE012-S2. PCIe and no problems booting off of it in 2008.11. On Jun 27, 2009, at 3:02 PM, Simon Breden no-re...@opensolaris.org wrote: Hi, Does anyone know of a reliable 2 or 4 port SATA card with a solid driver, that plugs into a PCIe slot, so that I can benefit from the high read speeds available from adding a couple of SSDs to form my ZFS root/boot pool? (Each SSD is capable of reading at around 150-200 MBytes/sec) After initially thinking I would move my existing 6-drive RAID-Z2 array to a new 8-port SATA controller, I finally decided to leave the drives connected to the motherboard SATA ports, and instead to get an additional smaller SATA card to allow me to connect 2 boot drives to form a mirror. For anyone considering a controller card to support 8 SATA drives, see this thread which has got some great comments from people experienced with using these larger cards. No doubt I will refer to it again when I build another storage system one day :) See: http://www.opensolaris.org/jive/thread.jspa?threadID=106210 Thanks, Simon http://breden.org.uk/2008/03/02/a-home-fileserver-using-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 mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] zpool import: Cannot mount,
Hi a small addendum. It seems that all sub ZFS below /atlashome/BACKUP are already mounted when /atlashome/BACKUP is tried to be mounted: # zfs get all atlashome/BACKUP|head -15 NAME PROPERTY VALUE SOURCE atlashome/BACKUP type filesystem - atlashome/BACKUP creation Thu Oct 9 16:30 2008 - atlashome/BACKUP used 9.95T - atlashome/BACKUP available 1.78T - atlashome/BACKUP referenced 172K - atlashome/BACKUP compressratio 1.47x - atlashome/BACKUP mountedno - atlashome/BACKUP quota none default atlashome/BACKUP reservationnone default atlashome/BACKUP recordsize 32Kinherited from atlashome atlashome/BACKUP mountpoint /atlashome/BACKUP default atlashome/BACKUP sharenfs on inherited from atlashome atlashome/BACKUP checksum on default atlashome/BACKUP compressionon local while # ls -l /atlashome/BACKUP | wc -l 33 Is there any way to force zpool import to re-order that? I could delete all stuff under BACKUP, however given the size I don't really want to. Cheers Carsten ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS and Dinamic Stripe
Thank you Relling and et151817 for your answers. So just to end the post: Relling supouse the next situation: One zpool in Dinamic Stripe with two disk, one of 100MB and the second with 200MB if the spread is stochastic spreading of data across vdevs you will have the double of possibilities of save one chunk in the second disk than in the first, right? Thanks for your time (and sorry for my english). JLBG -- 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 import: Cannot mount,
On Mon, 29 Jun 2009, Carsten Aulbert wrote: Is there any way to force zpool import to re-order that? I could delete all stuff under BACKUP, however given the size I don't really want to. Do a zpool export first, and then check to see what's in /atlashome. My bet is that the BACKUP directory is still there. If so, do an rmdir on /atlashome/BACKUP and then try the import again. Regards, markm ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Best controller card for 2 to 4 SATA drives ?
On Mon, Jun 29 at 11:43, Patrick O'Sullivan wrote: I've had success with the SIIG SC-SAE012-S2. PCIe and no problems booting off of it in 2008.11. I think there's a 4-port version of the 1068e-based chips from LSI, and I believe this is it: http://www.lsi.com/storage_home/products_home/host_bus_adapters/sas_hbas/lsisas3041er/index.html but I don't have one on-hand to confirm. The product brief states the LSISAS3041E-R leverages the LSISAS1064E controller ASIC's advanced Fusion-MPT architecture. so it appears to be the right board. Froogle lists it for about $150 from various websites. --eric -- Eric D. Mudama edmud...@mail.bounceswoosh.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] slow ls or slow zfs
On Fri, 26 Jun 2009, Richard Elling wrote: All the tools I have used show no IO problems. I think the problem is memory but I am unsure on how to troubleshoot it. Look for latency, not bandwidth. iostat will show latency at the device level. Unfortunately, the effect may not be all that obvious since the disks will only be driven as hard as the slowest disk and so the slowest disk may not seem much slower. Bob -- Bob Friesenhahn bfrie...@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discu ss I checked the output of iostat. svc_t is between 5 and 50, depending on when data is flushed to the disk (CIFS write pattern). %b is between 10 and 50. %w is always 0. Example: devicer/sw/s kr/s kw/s wait actv svc_t %w %b sd27 31.5 127.0 935.9 616.7 0.0 11.9 75.2 0 66 sd28 5.00.0 320.00.0 0.0 0.1 18.0 0 9 This tells me disks are busy but I do not know what they are doing? are they spending time seeking, writting or reading? I also review some ARC stats. Here is the output. ARC Efficency: Cache Access Total: 199758875 Cache Hit Ratio: 74% 148652045 [Defined State for buffer] Cache Miss Ratio: 25% 51106830 [Undefined State for Buffer] REAL Hit Ratio: 73% 146091795 [MRU/MFU Hits Only] Data Demand Efficiency:94% Data Prefetch Efficiency:15% CACHE HITS BY CACHE LIST: Anon: --%Counter Rolled. Most Recently Used: 22%33843327 (mru) [ Return Customer ] Most Frequently Used: 75%112248468 (mfu)[ Frequent Customer ] Most Recently Used Ghost:3%4833189 (mru_ghost)[ Return Customer Evicted, Now Back ] Most Frequently Used Ghost: 22%33831706 (mfu_ghost) [ Frequent Customer Evicted, Now Back ] It seems to me that mfu_ghost being at 22%, I may need a bigger ARC. Is ARC also designed to work with large memory foot prints (128GB for example or higher)? Will it be as efficient? -- 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] slow ls or slow zfs
NightBird wrote: On Fri, 26 Jun 2009, Richard Elling wrote: All the tools I have used show no IO problems. I think the problem is memory but I am unsure on how to troubleshoot it. Look for latency, not bandwidth. iostat will show latency at the device level. Unfortunately, the effect may not be all that obvious since the disks will only be driven as hard as the slowest disk and so the slowest disk may not seem much slower. Bob -- Bob Friesenhahn bfrie...@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discu ss I checked the output of iostat. svc_t is between 5 and 50, depending on when data is flushed to the disk (CIFS write pattern). %b is between 10 and 50. %w is always 0. Example: devicer/sw/s kr/s kw/s wait actv svc_t %w %b sd27 31.5 127.0 935.9 616.7 0.0 11.9 75.2 0 66 This is a slow disk. Put your efforts here. sd28 5.00.0 320.00.0 0.0 0.1 18.0 0 9 This tells me disks are busy but I do not know what they are doing? are they spending time seeking, writting or reading? I also review some ARC stats. Here is the output. ARC Efficency: Cache Access Total: 199758875 Cache Hit Ratio: 74% 148652045 [Defined State for buffer] Cache Miss Ratio: 25% 51106830 [Undefined State for Buffer] REAL Hit Ratio: 73% 146091795 [MRU/MFU Hits Only] Data Demand Efficiency:94% Data Prefetch Efficiency:15% CACHE HITS BY CACHE LIST: Anon: --%Counter Rolled. That is interesting... but only from a developer standpoint. Most Recently Used: 22%33843327 (mru) [ Return Customer ] Most Frequently Used: 75%112248468 (mfu)[ Frequent Customer ] Most Recently Used Ghost:3%4833189 (mru_ghost)[ Return Customer Evicted, Now Back ] Most Frequently Used Ghost: 22%33831706 (mfu_ghost) [ Frequent Customer Evicted, Now Back ] It seems to me that mfu_ghost being at 22%, I may need a bigger ARC. Is ARC also designed to work with large memory foot prints (128GB for example or higher)? Will it be as efficient? Caching isn't your problem, though adding memory may hide the real problem for a while. You need faster disk. -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] slow ls or slow zfs
On Mon, 29 Jun 2009, NightBird wrote: I checked the output of iostat. svc_t is between 5 and 50, depending on when data is flushed to the disk (CIFS write pattern). %b is between 10 and 50. %w is always 0. Example: devicer/sw/s kr/s kw/s wait actv svc_t %w %b sd27 31.5 127.0 935.9 616.7 0.0 11.9 75.2 0 66 sd28 5.00.0 320.00.0 0.0 0.1 18.0 0 9 This tells me disks are busy but I do not know what they are doing? are they spending time seeking, writting or reading? It looks like your sd27 is being pounded with write iops. It is close to its limit. Can you post complete iostat output? Since you have so many disks, (which may not always be involved in the same stripe) you may need to have iostat average over a long period of time such as 30 or 60 seconds in order to see a less responsive disk. Disks could be less responsive for many reasons, including vibrations in their operating environment. Also see Jeff Bonwick's diskqual.sh as described at http://www.mail-archive.com/zfs-discuss@opensolaris.org/msg15384.html which is good at helping to find pokey disks. A slightly modified version is included below. Bob -- Bob Friesenhahn bfrie...@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer,http://www.GraphicsMagick.org/ #!/bin/ksh # Date: Mon, 14 Apr 2008 15:49:41 -0700 # From: Jeff Bonwick jeff.bonw...@sun.com # To: Henrik Hjort hj...@dhs.nu # Cc: zfs-discuss@opensolaris.org # Subject: Re: [zfs-discuss] Performance of one single 'cp' # # No, that is definitely not expected. # # One thing that can hose you is having a single disk that performs # really badly. I've seen disks as slow as 5 MB/sec due to vibration, # bad sectors, etc. To see if you have such a disk, try my diskqual.sh # script (below). On my desktop system, which has 8 drives, I get: # # # ./diskqual.sh # c1t0d0 65 MB/sec # c1t1d0 63 MB/sec # c2t0d0 59 MB/sec # c2t1d0 63 MB/sec # c3t0d0 60 MB/sec # c3t1d0 57 MB/sec # c4t0d0 61 MB/sec # c4t1d0 61 MB/sec # # The diskqual test is non-destructive (it only does reads), but to # get valid numbers you should run it on an otherwise idle system. disks=`format /dev/null | grep ' c.t' | nawk '{print $2}'` getspeed1() { ptime dd if=/dev/rdsk/${1}s0 of=/dev/null bs=64k count=1024 21 | nawk '$1 == real { printf(%.0f\n, 67.108864 / $2) }' } getspeed() { # Best out of 6 for iter in 1 2 3 4 5 6 do getspeed1 $1 done | sort -n | tail -2 | head -1 } for disk in $disks do echo $disk `getspeed $disk` MB/sec done ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS and Dinamic Stripe
Jose Luis Barquín Guerola wrote: Thank you Relling and et151817 for your answers. So just to end the post: Relling supouse the next situation: One zpool in Dinamic Stripe with two disk, one of 100MB and the second with 200MB if the spread is stochastic spreading of data across vdevs you will have the double of possibilities of save one chunk in the second disk than in the first, right? The simple answer is yes. The more complex answer is that copies will try to be spread across different vdevs. Metadata, by default, uses copies=2, so you could expect the metadata to be more evenly spread across the disks. -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] slow ls or slow zfs
On Mon, 29 Jun 2009, NightBird wrote: I checked the output of iostat. svc_t is between 5 and 50, depending on when data is flushed to the disk (CIFS write pattern). %b is between 10 and 50. %w is always 0. Example: devicer/sw/s kr/s kw/s wait actv svc_t %w %b sd27 31.5 127.0 935.9 616.7 0.0 11.9 75.2 0 66 d28 5.00.0 320.00.0 0.0 0.1 18.0 0 9 This tells me disks are busy but I do not know what they are doing? are they spending time seeking, writting or reading? It looks like your sd27 is being pounded with write iops. It is close to its limit. Can you post complete iostat output? Since you have so many disks, (which may not always be involved in the same stripe) you may need to have iostat average over a long period of time such as 30 or 60 seconds in order to see a less responsive disk. Disks could be less esponsive for many reasons, including vibrations in their operating environment. Also see Jeff Bonwick's diskqual.sh as described at http://www.mail-archive.com/zfs-disc...@opensolaris.or g/msg15384.html which is good at helping to find pokey disks. A slightly modified version is included below. Bob -- Bob Friesenhahn bfrie...@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ I will run the script when the server is idle as recommended and report back. Here is the full iostat output (30sec). c9t40d0 seems to have a consistently higher svc_t time. r/sw/s kr/s kw/s wait actv wsvc_t asvc_t %w %b device 0.02.62.1 11.0 0.0 0.00.06.5 0 1 c8t0d0 0.12.60.4 11.0 0.0 0.00.05.1 0 0 c8t1d0 6.06.2 380.1 147.8 0.0 0.30.0 23.6 0 13 c9t8d0 6.26.5 390.8 147.9 0.0 0.30.0 21.9 0 13 c9t9d0 6.36.2 386.1 147.6 0.0 0.30.0 26.7 0 12 c9t10d0 6.76.2 413.5 147.8 0.0 0.30.0 21.9 0 14 c9t11d0 6.15.7 371.1 147.6 0.0 0.30.0 21.2 0 11 c9t12d0 6.75.9 407.3 147.6 0.0 0.30.0 21.4 0 13 c9t13d0 5.76.3 347.6 147.7 0.0 0.30.0 22.4 0 12 c9t14d0 7.05.9 426.5 147.5 0.0 0.30.0 20.6 0 13 c9t15d0 6.66.1 405.0 147.6 0.0 0.30.0 21.1 0 12 c9t16d0 6.66.2 405.2 147.7 0.0 0.30.0 21.1 0 12 c9t17d0 7.16.3 432.9 147.8 0.0 0.30.0 20.9 0 14 c9t18d0 6.76.5 411.6 147.9 0.0 0.30.0 23.6 0 13 c9t19d0 6.46.4 390.3 148.1 0.0 0.30.0 21.7 0 13 c9t20d0 6.96.9 424.4 147.9 0.0 0.30.0 19.8 0 13 c9t21d0 6.26.9 375.3 148.1 0.0 0.30.0 20.2 0 12 c9t22d0 5.76.8 349.5 147.9 0.0 0.30.0 20.9 0 12 c9t23d0 6.26.6 377.5 147.9 0.0 0.30.0 20.6 0 11 c9t24d0 5.46.7 328.2 147.9 0.0 0.20.0 20.7 0 11 c9t25d0 6.76.7 407.3 148.0 0.0 0.30.0 19.8 0 12 c9t26d0 6.56.9 396.7 148.1 0.0 0.30.0 20.4 0 13 c9t27d0 6.46.6 390.4 147.9 0.0 0.30.0 21.3 0 13 c9t28d0 6.86.3 416.0 147.6 0.0 0.40.0 26.8 0 13 c9t29d0 6.86.3 413.9 147.8 0.0 0.30.0 23.5 0 13 c9t30d0 7.5 33.5 446.9 312.0 0.0 1.80.0 45.0 0 18 c9t31d0 8.2 33.6 491.7 312.0 0.0 2.10.0 51.1 0 21 c9t32d0 7.0 34.3 414.9 312.3 0.0 1.90.0 47.0 0 20 c9t33d0 7.6 34.1 463.4 312.2 0.0 2.10.0 51.2 0 21 c9t34d0 7.9 33.5 474.4 312.0 0.0 2.20.0 52.9 0 21 c9t35d0 8.2 33.2 496.0 311.7 0.0 2.40.0 59.1 0 23 c9t36d0 8.0 33.2 481.0 311.9 0.0 2.00.0 48.8 0 21 c9t37d0 7.8 33.4 469.9 311.9 0.0 2.30.0 56.4 0 20 c9t38d0 8.5 34.1 518.7 312.4 0.0 2.30.0 54.3 0 22 c9t39d0 8.4 32.9 510.5 311.8 0.0 2.90.0 70.6 0 27 c9t40d0 8.2 34.3 501.5 312.4 0.0 2.30.0 55.1 0 24 c9t41d0 8.1 34.3 491.1 312.5 0.0 2.30.0 55.4 0 21 c9t42d0 8.5 34.3 510.9 312.7 0.0 2.30.0 53.3 0 23 c9t43d0 7.5 34.3 453.1 312.6 0.0 2.30.0 54.4 0 20 c9t44d0 7.0 33.7 420.9 312.1 0.0 2.30.0 55.7 0 19 c9t45d0 7.0 34.2 420.9 312.4 0.0 2.30.0 55.2 0 20 c9t46d0 7.9 35.1 474.6 312.5 0.0 2.10.0 49.1 0 22 c9t47d0 8.1 35.0 487.4 312.8 0.0 2.30.0 52.8 0 22 c9t48d0 8.1 34.2 491.3 312.2 0.0 2.10.0 50.2 0 20 c9t49d0 7.2 34.6 429.4 312.5 0.0 2.10.0 51.3 0 20 c9t50d0 7.6 35.2 459.3 312.6 0.0 2.30.0 54.1 0 21 c9t51d0 7.7 35.0 463.5 312.6 0.0 2.10.0 49.2 0 21 c9t52d0 7.4 35.1 442.6 312.8 0.0 2.10.0 48.8 0 21 c9t53d0 -- This message posted from
Re: [zfs-discuss] zpool import: Cannot mount,
Hi Mark J Musante wrote: Do a zpool export first, and then check to see what's in /atlashome. My bet is that the BACKUP directory is still there. If so, do an rmdir on /atlashome/BACKUP and then try the import again. Sorry, I meant to copy this earlier: s11 console login: root Password: Last login: Mon Jun 29 10:37:47 on console Sun Microsystems Inc. SunOS 5.10 Generic January 2005 s11:~# zpool export atlashome s11:~# ls -l /atlashome /atlashome: No such file or directory s11:~# zpool import atlashome cannot mount '/atlashome/BACKUP': directory is not empty s11:~# ls -l /atlashome/BACKUP/|wc -l 33 s11:~# Thus you see that probably zpool import does the wrong thing(TM) (or wrong order) Any idea? Cheers Carsten PS: I opened a case for that, but waited for the call back. When solving this problem, I can post the case ID for further reference. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Cloning a ZFS compact flash card
On 06/28/09 08:41, Ross wrote: Can't you just boot from an OpenSolaris CD, create a ZFS pool on the new device, and just do a ZFS send/receive directly to it? So long as there's enough space for the data, a send/receive won't care at all that the systems are different sizes. I don't know what you need to do to a pool to make it bootable I'm afraid, so I don't know if this will just work or if it'll need some more tweaking. However if you have any problems you should be able to find more information about how ZFS boot works online. A procedure for making a backup of a root pool and then restoring it is at: http://www.solarisinternals.com/wiki/index.php/ZFS_Troubleshooting_Guide#ZFS_Root_Pool_Recovery Lori ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] zpool import: Cannot mount,
On Mon, 29 Jun 2009, Carsten Aulbert wrote: s11 console login: root Password: Last login: Mon Jun 29 10:37:47 on console Sun Microsystems Inc. SunOS 5.10 Generic January 2005 s11:~# zpool export atlashome s11:~# ls -l /atlashome /atlashome: No such file or directory s11:~# zpool import atlashome cannot mount '/atlashome/BACKUP': directory is not empty s11:~# ls -l /atlashome/BACKUP/|wc -l 33 s11:~# OK, looks like you're running into CR 6827199. There's a workaround for that as well. After the zpool import, manually zfs umount all the datasets under /atlashome/BACKUP. Once you've done that, the BACKUP directory will still be there. Manually mount the dataset that corresponds to /atlashome/BACKUP, and then try 'zfs mount -a'. Regards, markm ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] zpool import: Cannot mount,
Hi Mark, Mark J Musante wrote: OK, looks like you're running into CR 6827199. There's a workaround for that as well. After the zpool import, manually zfs umount all the datasets under /atlashome/BACKUP. Once you've done that, the BACKUP directory will still be there. Manually mount the dataset that corresponds to /atlashome/BACKUP, and then try 'zfs mount -a'. I did that (needed to rmdir the directories under BACKUP) and then finally it worked - and the best even after a reboot it was able to mount all file systems again. Great and a lot of thanks! One question: Where can I find more about CR 6827199? I logged into sun.com with my service contract enabled log-in but I cannot find it there (or the search function does not like me too much). Cheers Carsten ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] zpool import: Cannot mount,
On 29.06.09 23:01, Carsten Aulbert wrote: One question: Where can I find more about CR 6827199? I logged into sun.com with my service contract enabled log-in but I cannot find it there (or the search function does not like me too much). You can try bugs.opensolaris.org too: http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6827199 victor ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS - SWAP and lucreate..
Hi Patrick, To answer your original question, yes, you can create your root swap and dump volumes before you run the lucreate operation. LU won't change them if they are already created. Keep in mind that you'll need approximately 10 GBs of disk space for the ZFS root BE and the swap/dump volumes. See the steps below. Cindy Patrick Bittner wrote: So there is no possibility to do this with or before the lucreate command? hm. well- thank you anyway then # zpool create rpool c0t0d0s0 # zfs create -V 1g rpool/dump # zfs create -V 2g -b 8k rpool/swap # zfs list NAME USED AVAIL REFER MOUNTPOINT rpool 3.00G 30.2G18K /rpool rpool/dump 1G 31.2G16K - rpool/swap 2G 32.2G16K - # lucreate -c ufsBE -n zfsBE -p rpool . . . . Population of boot environment zfsBE successful. Creation of boot environment zfsBE successful. # zfs list NAME USED AVAIL REFER MOUNTPOINT rpool 7.62G 25.6G 92.5K /rpool rpool/ROOT4.62G 25.6G18K /rpool/ROOT rpool/ROOT/zfsBE 4.62G 25.6G 4.62G / rpool/dump 1G 26.6G16K - rpool/swap 2G 27.6G16K - ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] L2ARC availability w/ Solaris 10
Hi, Is there a time frame when L2ARC would be available in Solaris 10. With the latest U7 release, L2ARC appears to be disabled (operation not supported on this type of pool). Thanks, Ravi -- This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] L2ARC availability w/ Solaris 10
Hi, Is there a time frame when L2ARC would be available in Solaris 10. With the latest U7 release, L2ARC appears to be disabled (operation not supported on this type of pool). Thanks, Ravi -- *Ravi Kota ISV Engineering Sun Microsystems, Inc. Phone: 408-228-1264, x69401 Mobile: 408-393-3624 Email: ravi.k...@sun.com mailto:ravi.k...@sun.com* ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS and Dinamic Stripe
try to be spread across different vdevs. % zpool iostat -v capacity operationsbandwidth pool used avail read write read write -- - - - - - - z686G 434G 40 5 2.46M 271K c1t0d0s7 250G 194G 14 1 877K 94.2K c1t1d0s7 244G 200G 15 2 948K 96.5K c0d0 193G 39.1G 10 1 689K 80.2K note that c0d0 is basically full, but still serving 10 of every 15 reads, and 82% of the writes. Rob ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] SXCE, ZFS root, b101 - b103 fails with ERROR: No upgradeable file systems
On 06/27/09 23:50, Ian Collins wrote: Leela wrote: So no one has any idea? About what? This was in regards to a question sent to the install-discuss alias on 6/18 and later copied to zfs-discuss. I have answered it on the install alias, if anyone is following the issue. Lori ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS write I/O stalls
On Wed, 24 Jun 2009, Lejun Zhu wrote: There is a bug in the database about reads blocked by writes which may be related: http://bugs.opensolaris.org/view_bug.do?bug_id=6471212 The symptom is sometimes reducing queue depth makes read perform better. I have been banging away at this issue without resolution. Based on Roch Bourbonnais's blog description of the ZFS write throttle code, it seems that I am facing a perfect storm. Both the storage write bandwidth (800+ MB/second) and the memory size of my system (20 GB) result in the algorithm batching up 2.5 GB of user data to write. Since I am using mirrors, this results in 5 GB of data being written at full speed to the array on a very precise schedule since my application is processing fixed-sized files with a fixed algorithm. The huge writes lead to at least 3 seconds of read starvation, resulting in a stalled application and a square-wave of system CPU utilization. I could attempt to modify my application to read ahead by 3 seconds but that would require gigabytes of memory, lots of complexity, and would not be efficient. Richard Elling thinks that my array is pokey, but based on write speed and memory size, ZFS is always going to be batching up data to fill the write channel for 5 seconds so it does not really matter how fast that write channel is. If I had 32GB of RAM and 2X the write speed, the situation would be identical. Hopefully someone at Sun is indeed working this read starvation issue and it will be resolved soon. Bob -- Bob Friesenhahn bfrie...@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer,http://www.GraphicsMagick.org/ ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS write I/O stalls
On Mon, Jun 29, 2009 at 2:48 PM, Bob Friesenhahnbfrie...@simple.dallas.tx.us wrote: On Wed, 24 Jun 2009, Lejun Zhu wrote: There is a bug in the database about reads blocked by writes which may be related: http://bugs.opensolaris.org/view_bug.do?bug_id=6471212 The symptom is sometimes reducing queue depth makes read perform better. I have been banging away at this issue without resolution. Based on Roch Bourbonnais's blog description of the ZFS write throttle code, it seems that I am facing a perfect storm. Both the storage write bandwidth (800+ MB/second) and the memory size of my system (20 GB) result in the algorithm batching up 2.5 GB of user data to write. Since I am using mirrors, this results in 5 GB of data being written at full speed to the array on a very precise schedule since my application is processing fixed-sized files with a fixed algorithm. The huge writes lead to at least 3 seconds of read starvation, resulting in a stalled application and a square-wave of system CPU utilization. I could attempt to modify my application to read ahead by 3 seconds but that would require gigabytes of memory, lots of complexity, and would not be efficient. Richard Elling thinks that my array is pokey, but based on write speed and memory size, ZFS is always going to be batching up data to fill the write channel for 5 seconds so it does not really matter how fast that write channel is. If I had 32GB of RAM and 2X the write speed, the situation would be identical. Hopefully someone at Sun is indeed working this read starvation issue and it will be resolved soon. Bob -- Bob Friesenhahn bfrie...@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss I see similar square-wave performance. However, my load is primarily write-based, when those commits happen, I see all network activity pause while the buffer is commited to disk. I write about 750Mbit/sec over the network to the X4540's during backup windows using primarily iSCSI. When those writes occur to my RaidZ volume, all activity pauses until the writes are fully flushed. One thing to note, on 117, the effects are seemingly reduced and a bit more even performance, but it is still there. -- Brent Jones br...@servuhome.net ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS write I/O stalls
On Wed, 24 Jun 2009, Lejun Zhu wrote: There is a bug in the database about reads blocked by writes which may be related: http://bugs.opensolaris.org/view_bug.do?bug_id=6471212 The symptom is sometimes reducing queue depth makes read perform better. I have been banging away at this issue without resolution. Based on Roch Bourbonnais's blog description of the ZFS write throttle code, it seems that I am facing a perfect storm. Both the storage write bandwidth (800+ MB/second) and the memory size of my system (20 GB) result in the algorithm batching up 2.5 GB of user data to write. With ZFS write throttle, the number 2.5GB is tunable. From what I've read in the code, it is possible to e.g. set zfs:zfs_write_limit_override = 0x800 (bytes) to make it write 128M instead. Since I am using mirrors, this results in 5 GB of data being written at full speed to the array on a very precise schedule since my application is processing fixed-sized files with a fixed algorithm. The huge writes lead to at least 3 seconds of read starvation, resulting in a stalled application and a square-wave of system CPU utilization. I could attempt to modify my application to read ahead by 3 seconds but that would require gigabytes of memory, lots of complexity, and would not be efficient. Richard Elling thinks that my array is pokey, but based on write speed and memory size, ZFS is always going to be batching up data to fill the write channel for 5 seconds so it does not really matter how fast that write channel is. If I had 32GB of RAM and 2X the write speed, the situation would be identical. Hopefully someone at Sun is indeed working this read starvation issue and it will be resolved soon. Bob -- Bob Friesenhahn bfrie...@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discu ss -- This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss