Re: [zfs-discuss] Improving L1ARC cache efficiency with dedup
On 12/07/11 20:48, Mertol Ozyoney wrote: Unfortunetly the answer is no. Neither l1 nor l2 cache is dedup aware. The only vendor i know that can do this is Netapp In fact , most of our functions, like replication is not dedup aware. For example, thecnicaly it's possible to optimize our replication that it does not send daya chunks if a data chunk with the same chechsum exists in target, without enabling dedup on target and source. We already do that with 'zfs send -D': -D Perform dedup processing on the stream. Deduplicated streams cannot be received on systems that do not support the stream deduplication feature. -- Darren J Moffat ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Improving L1ARC cache efficiency with dedup
> From: zfs-discuss-boun...@opensolaris.org [mailto:zfs-discuss- > boun...@opensolaris.org] On Behalf Of Mertol Ozyoney > Sent: Wednesday, December 07, 2011 3:49 PM > To: Brad Diggs > Cc: zfs-discuss@opensolaris.org > Subject: Re: [zfs-discuss] Improving L1ARC cache efficiency with dedup > > Unfortunetly the answer is no. Neither l1 nor l2 cache is dedup aware. I haven't read the code, but I can reference experimental results that seem to defy statement... If you time write a large data stream of completely duplicated data to disk without dedup... And time read it back... It takes the same amount of time. But, If you enable dedup and repeat the same test, it goes much faster. Depending on a lot of variables, it might be 2x-12x faster. To me, "significantly faster than disk speed," can only mean it's benefitting from cache. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Issues with Areca 1680
Hi all, I have a server that is build on top of an Asus board which is equipped with an Areca 1680 HBA. Since ZFS like raw disks, I changed its mode from RAID to JBOD in the firmware and rebootet the host. Now, I do have 16 drives in the chassis and the line out like this: root@vsm01:~# format Searching for disks...done c5t0d0: configured with capacity of 978.00MB AVAILABLE DISK SELECTIONS: 0. c3t1d0 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,0 1. c3t1d1 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,1 2. c3t1d2 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,2 3. c3t1d3 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,3 4. c3t1d4 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,4 5. c3t1d5 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,5 6. c3t1d6 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,6 7. c3t1d7 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,7 8. c3t2d0 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,0 9. c3t2d1 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,1 10. c3t2d2 sec 126> /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,2 11. c3t2d3 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,3 12. c3t2d4 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,4 13. c3t2d5 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,5 14. c3t2d6 sec 126> /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,6 15. c3t2d7 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,7 16. c5t0d0 /pci@0,0/pci1043,819e@1d,7/storage@4/disk@0,0 Looked okay for me, so I went ahead an created a zpool containing two mirrors like this: root@vsm01:~# zpool create vsm_pool1_1T mirror c3t2d1 c3t1d6 mirror c3t2d0 c3t1d5 This went just fine and the zpool was created like this: root@vsm01:~# zpool status vsm_pool1_1T pool: vsm_pool1_1T state: ONLINE scan: none requested config: NAMESTATE READ WRITE CKSUM vsm_pool1_1T ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 c3t2d1 ONLINE 0 0 0 c3t1d6 ONLINE 0 0 0 mirror-1 ONLINE 0 0 0 c3t2d0 ONLINE 0 0 0 c3t1d5 ONLINE 0 0 0 errors: No known data errors Now, creating another zpool from the remaining 500GB drives failed with this weird error: root@vsm01:~# zpool create vsm_pool2_1T mirror c3t1d4 c3t1d3mirror c3t1d2 c3t1d1 invalid vdev specification use '-f' to override the following errors: /dev/dsk/c3t2d1s0 is part of active ZFS pool vsm_pool1_1T. Please see zpool(1M). Anybody has an idea of what is going wrong here. It doesn't seem to matter which of the drives I want to use for the new zpool, I am always getting this error message - ven when trying only mirror c3t1d4 c3t1d3 or mirror c3t1d2 c3t1d1 alone. Thanks, budy ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] SOLVED: Issues with Areca 1680
Am 08.12.11 18:14, schrieb Stephan Budach: Hi all, I have a server that is build on top of an Asus board which is equipped with an Areca 1680 HBA. Since ZFS like raw disks, I changed its mode from RAID to JBOD in the firmware and rebootet the host. Now, I do have 16 drives in the chassis and the line out like this: root@vsm01:~# format Searching for disks...done c5t0d0: configured with capacity of 978.00MB AVAILABLE DISK SELECTIONS: 0. c3t1d0 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,0 1. c3t1d1 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,1 2. c3t1d2 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,2 3. c3t1d3 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,3 4. c3t1d4 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,4 5. c3t1d5 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,5 6. c3t1d6 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,6 7. c3t1d7 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@1,7 8. c3t2d0 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,0 9. c3t2d1 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,1 10. c3t2d2 sec 126> /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,2 11. c3t2d3 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,3 12. c3t2d4 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,4 13. c3t2d5 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,5 14. c3t2d6 sec 126> /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,6 15. c3t2d7 /pci@0,0/pci8086,29f1@1/pci17d3,1680@0/disk@2,7 16. c5t0d0 /pci@0,0/pci1043,819e@1d,7/storage@4/disk@0,0 Looked okay for me, so I went ahead an created a zpool containing two mirrors like this: root@vsm01:~# zpool create vsm_pool1_1T mirror c3t2d1 c3t1d6 mirror c3t2d0 c3t1d5 This went just fine and the zpool was created like this: root@vsm01:~# zpool status vsm_pool1_1T pool: vsm_pool1_1T state: ONLINE scan: none requested config: NAMESTATE READ WRITE CKSUM vsm_pool1_1T ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 c3t2d1 ONLINE 0 0 0 c3t1d6 ONLINE 0 0 0 mirror-1 ONLINE 0 0 0 c3t2d0 ONLINE 0 0 0 c3t1d5 ONLINE 0 0 0 errors: No known data errors Now, creating another zpool from the remaining 500GB drives failed with this weird error: root@vsm01:~# zpool create vsm_pool2_1T mirror c3t1d4 c3t1d3mirror c3t1d2 c3t1d1 invalid vdev specification use '-f' to override the following errors: /dev/dsk/c3t2d1s0 is part of active ZFS pool vsm_pool1_1T. Please see zpool(1M). Anybody has an idea of what is going wrong here. It doesn't seem to matter which of the drives I want to use for the new zpool, I am always getting this error message - ven when trying only mirror c3t1d4 c3t1d3 or mirror c3t1d2 c3t1d1 alone. Thanks, budy Hmm… now I did it the other way round and this time it worked as expected. Seems that these disks have been used in some other zpool configurations without being properly removed and thus they may had some old labels on them, right? Thanks ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Improving L1ARC cache efficiency with dedup
On 12/ 9/11 12:39 AM, Darren J Moffat wrote: On 12/07/11 20:48, Mertol Ozyoney wrote: Unfortunetly the answer is no. Neither l1 nor l2 cache is dedup aware. The only vendor i know that can do this is Netapp In fact , most of our functions, like replication is not dedup aware. For example, thecnicaly it's possible to optimize our replication that it does not send daya chunks if a data chunk with the same chechsum exists in target, without enabling dedup on target and source. We already do that with 'zfs send -D': -D Perform dedup processing on the stream. Deduplicated streams cannot be received on systems that do not support the stream deduplication feature. Is there any more published information on how this feature works? -- Ian. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Improving L1ARC cache efficiency with dedup
You can see the original ARC case here: http://arc.opensolaris.org/caselog/PSARC/2009/557/20091013_lori.alt On 8 Dec 2011, at 16:41, Ian Collins wrote: > On 12/ 9/11 12:39 AM, Darren J Moffat wrote: >> On 12/07/11 20:48, Mertol Ozyoney wrote: >>> Unfortunetly the answer is no. Neither l1 nor l2 cache is dedup aware. >>> >>> The only vendor i know that can do this is Netapp >>> >>> In fact , most of our functions, like replication is not dedup aware. >>> For example, thecnicaly it's possible to optimize our replication that >>> it does not send daya chunks if a data chunk with the same chechsum >>> exists in target, without enabling dedup on target and source. >> We already do that with 'zfs send -D': >> >> -D >> >> Perform dedup processing on the stream. Deduplicated >> streams cannot be received on systems that do not >> support the stream deduplication feature. >> >> >> >> > Is there any more published information on how this feature works? > > -- > Ian. > > ___ > 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] First zone creation - getting ZFS error
I would also try it without the /zones mountpoint. Putting the zone root dir on an alternate mountpoint caused problems for us. Try creating /datastore/zones for a zone root home, or just make the zones in /datastore Solaris seems to get very easily confused when zone root is anything out of the ordinary ( and it really bites you at patch time!) On Dec 7, 2011, at 9:50 PM, Ian Collins wrote: > On 12/ 7/11 05:12 AM, Mark Creamer wrote: >> I'm running OI 151a. I'm trying to create a zone for the first time, and am >> getting an error about zfs. I'm logged in as me, then su - to root before >> running these commands. >> >> I have a pool called datastore, mounted at /datastore >> >> Per the wiki document http://wiki.openindiana.org/oi/Building+in+zones, I >> first created the zfs file system (note that the command syntax in the >> document appears to be wrong, so I did the options I wanted separately): >> >> zfs create datastore/zones >> zfs set compression=on datastore/zones >> zfs set mountpoint=/zones datastore/zones >> >> zfs list shows: >> >> NAME USED AVAIL REFER MOUNTPOINT >> datastore 28.5M 7.13T 57.9K /datastore >> datastore/dbdata28.1M 7.13T 28.1M /datastore/dbdata >> datastore/zones 55.9K 7.13T 55.9K /zones >> rpool 27.6G 201G45K /rpool >> rpool/ROOT 2.89G 201G31K legacy >> rpool/ROOT/openindiana 2.89G 201G 2.86G / >> rpool/dump 12.0G 201G 12.0G - >> rpool/export5.53M 201G32K /export >> rpool/export/home 5.50M 201G32K /export/home >> rpool/export/home/mcreamer 5.47M 201G 5.47M /export/home/mcreamer >> rpool/swap 12.8G 213G 137M - >> >> Then I went about creating the zone: >> >> zonecfg -z zonemaster >> create >> set autoboot=true >> set zonepath=/zones/zonemaster >> set ip-type=exclusive >> add net >> set physical=vnic0 >> end >> exit >> >> That all goes fine, then... >> >> zoneadm -z zonemaster install >> >> which returns... >> >> ERROR: the zonepath must be a ZFS dataset. >> The parent directory of the zonepath must be a ZFS dataset so that the >> zonepath ZFS dataset can be created properly. >> > That's odd, it should have worked. > >> Since the zfs dataset datastore/zones is created, I don't understand what >> the error is trying to get me to do. Do I have to do: >> >> zfs create datastore/zones/zonemaster >> >> before I can create a zone in that path? That's not in the documentation, so >> I didn't want to do anything until someone can point out my error for me. >> Thanks for your help! >> > You shouldn't have to, but it won't do any harm. > > If you don't get any further, try zones-discuss. > > -- > Ian. > > ___ > 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] First zone creation - getting ZFS error
On 12/ 9/11 11:37 AM, Betsy Schwartz wrote: On Dec 7, 2011, at 9:50 PM, Ian Collins wrote: On 12/ 7/11 05:12 AM, Mark Creamer wrote: Since the zfs dataset datastore/zones is created, I don't understand what the error is trying to get me to do. Do I have to do: zfs create datastore/zones/zonemaster before I can create a zone in that path? That's not in the documentation, so I didn't want to do anything until someone can point out my error for me. Thanks for your help! You shouldn't have to, but it won't do any harm. If you don't get any further, try zones-discuss. I would also try it without the /zones mountpoint. Putting the zone root dir on an alternate mountpoint caused problems for us. Try creating /datastore/zones for a zone root home, or just make the zones in /datastore Solaris seems to get very easily confused when zone root is anything out of the ordinary ( and it really bites you at patch time!) It shouldn't. On all my systems, I have: NAME USED AVAIL REFER MOUNTPOINT rpool/zoneRoot 11.6G 214G40K /zoneRoot -- Ian. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Scrub found error in metadata:0x0, is that always fatal? No checks um errors now...
On Mon, Dec 5, 2011 at 17:46, Jim Klimov wrote: > So, in contrast with Nigel's optimistic theory that > metadata is anyway extra-redundant and should be > easily fixable, it seems that I do still have the > problem. It does not show itself in practice as of > yet, but is found by scrub ;) Hmm. Interesting. I have re-scrubbed pool that I referenced. I haven't gotten another error in the metadata. > After a few days to complete the current scrub, > I plan to run zdb as asked by Steve. If anyone else > has some theories, suggestions or requests to dig > up more clues - bring them on! ;) > Perhaps the cause of the corruption is still active. The circumstances that lead up to the discovery of the error are different for you and me. The server that I encountered it on was running fine for months, it was only after the crash/hang because of attempting to add a bad drive to the pool that I encountered the issue which was found on boot in my case. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss