Re: [zfs-discuss] surprisingly poor performance

2009-07-07 Thread James Andrewartha
James Lever wrote: > > On 07/07/2009, at 8:20 PM, James Andrewartha wrote: > >> Have you tried putting the slog on this controller, either as an SSD or >> regular disk? It's supported by the mega_sas driver, x86 and amd64 only. > > What exactly are you suggesting here? Configure one disk on thi

Re: [zfs-discuss] Losts of small files vs fewer big files

2009-07-07 Thread Miles Nordin
> "dt" == Don Turnbull writes: dt> Any idea why this is? maybe prefetch? WAG, though. dt> I work with Greenplum which is essentially a number of dt> Postgres database instances clustered together. haha, yeah I know who you are. Too bad the open source postgres can't do that.

[zfs-discuss] zpool import hangs

2009-07-07 Thread Nicholas
I am having trouble with a Raid-Z zpool "bigtank" of 5x 750GB drives that will not import. After having some trouble with this pool, I exported it and attempted a reimport only to discover this issue: I can see the pool by running zpool import, and the devices are all online however running "zp

Re: [zfs-discuss] Single disk parity

2009-07-07 Thread Daniel Carosone
> Sorry, don't have a thread reference > to hand just now. http://www.opensolaris.org/jive/thread.jspa?threadID=100296 Note that there's little empirical evidence that this is directly applicable to the kinds of errors (single bit, or otherwise) that a single failing disk medium would produce.

Re: [zfs-discuss] Single disk parity

2009-07-07 Thread Christian Auby
> Do you have data to back this up? It's more of a logical observation. The random data corruption I've had up through the years have generally either involved a single sector or two or a full disk failure. 5% parity on a 128KB block size would allow you to lose 6.4KB, or ~10 512 byte sectors.

Re: [zfs-discuss] Single disk parity

2009-07-07 Thread Richard Elling
Christian Auby wrote: You are describing the copies parameter. It really helps to describe it in pictures, rather than words. So I did that. http://blogs.sun.com/relling/entry/zfs_copies_and_data _protection -- richard It's not quite like copies as it's not actually a copy of the data I

Re: [zfs-discuss] Single disk parity

2009-07-07 Thread Christian Auby
> You are describing the copies parameter. It really > helps to describe > it in pictures, rather than words. So I did that. > http://blogs.sun.com/relling/entry/zfs_copies_and_data > _protection > -- richard It's not quite like copies as it's not actually a copy of the data I'm talking about.

Re: [zfs-discuss] Losts of small files vs fewer big files

2009-07-07 Thread Don Turnbull
Thanks for the suggestion! We've fiddled with this in the past. Our app is 32k instead of 8k blocks and it is data warehousing so the I/O model is a lot more long sequential reads generally. Changing the blocksize has very little effect on us. I'll have to look at fsync; hadn't considered t

Re: [zfs-discuss] Losts of small files vs fewer big files

2009-07-07 Thread Bryan Allen
Have you set the recordsize for the filesystem to the blocksize Postgres is using (8K)? Note this has to be done before any files are created. Other thoughts: Disable postgres's fsync, enable filesystem compression if disk I/O is your bottleneck as opposed to CPU. I do this with MySQL and it has p

Re: [zfs-discuss] Single disk parity

2009-07-07 Thread Daniel Carosone
There was a discussion in zfs-code around error-correcting (rather than just -detecting) properties of the checksums currently kept, an of potential additional checksum methods with stronger properties. It came out of another discussion about fletcher2 being both weaker than desired, and flawed

[zfs-discuss] Losts of small files vs fewer big files

2009-07-07 Thread Don Turnbull
I work with Greenplum which is essentially a number of Postgres database instances clustered together. Being postgres, the data is held in a lot of individual files which can be each fairly big (hundreds of MB or several GB) or very small (50MB or less). We've noticed a performance difference

Re: [zfs-discuss] Single disk parity

2009-07-07 Thread Louis-Frédéric Feuillette
On Tue, 2009-07-07 at 17:42 -0700, Richard Elling wrote: > Christian Auby wrote: > > ZFS is able to detect corruption thanks to checksumming, but for single > > drives (regular folk-pcs) it doesn't help much unless it can correct them. > > I've been searching and can't find anything on the topic,

Re: [zfs-discuss] [nfs-discuss] NFS, ZFS & ESX

2009-07-07 Thread Dai Ngo
Without any tuning, the default TCP window size and send buffer size for NFS connections is around 48KB which is not very optimal for bulk transfer. However the 1.4MB/s write seems to indicate something else is seriously wrong. iSCSI performance was good, so the network connection seems to be O

Re: [zfs-discuss] [nfs-discuss] NFS, ZFS & ESX

2009-07-07 Thread Calum Mackay
interesting; but presumably the ZIL/fsflush is not the reason for the associated poor *read* performance? where does latencytop point the finger in that case? cheers, calum. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensola

Re: [zfs-discuss] Single disk parity

2009-07-07 Thread Richard Elling
Christian Auby wrote: ZFS is able to detect corruption thanks to checksumming, but for single drives (regular folk-pcs) it doesn't help much unless it can correct them. I've been searching and can't find anything on the topic, so here goes: 1. Can ZFS do parity data on a single drive? e.g. x%

[zfs-discuss] Single disk parity

2009-07-07 Thread Christian Auby
ZFS is able to detect corruption thanks to checksumming, but for single drives (regular folk-pcs) it doesn't help much unless it can correct them. I've been searching and can't find anything on the topic, so here goes: 1. Can ZFS do parity data on a single drive? e.g. x% parity for all writes,

Re: [zfs-discuss] surprisingly poor performance

2009-07-07 Thread James Lever
On 07/07/2009, at 8:20 PM, James Andrewartha wrote: Have you tried putting the slog on this controller, either as an SSD or regular disk? It's supported by the mega_sas driver, x86 and amd64 only. What exactly are you suggesting here? Configure one disk on this array as a dedicated ZIL?

[zfs-discuss] new zilstat beta testers wanted

2009-07-07 Thread Richard Elling
I've changed zilstat to have the option of recording changes between txg commits in addition to the current time-based changes. Before I turn it loose in the wild, I'd like a few testers who have interesting sync workloads to try it out. If you are interested, sign up by responding to my blog ent

Re: [zfs-discuss] possible to override/inherit mountpoint on received snapshots?

2009-07-07 Thread Cindy . Swearingen
FYI... The -u option is described in the ZFS admin guide and the ZFS troubleshooting wiki in the areas of restoring root pool snapshots. The -u option is described in the zfs.1m man page starting in the b115 release: http://docs.sun.com/app/docs/doc/819-2240/zfs-1m Cindy Lori Alt wrote: T

Re: [zfs-discuss] recover data after zpool create

2009-07-07 Thread stephen bond
None of the file recovery tools work with zfs. Testdisk is most advanced and the author is looking at incorporating zfs, but when will it happen nobody knows. I want to try with dd. Can anybody give me an example of how to read bytes cylinder by cylinder? Filtering the output is easy and I wi

Re: [zfs-discuss] zfs scheduled replication script?

2009-07-07 Thread Mark McCoy
bump -- 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] possible to override/inherit mountpoint on received snapshots?

2009-07-07 Thread Lori Alt
To elaborate, the -u option to zfs receive suppresses all mounts. The datasets you extract will STILL have mountpoints that might not work on the local system, but at least you can unpack the entire hierarchy of datasets and then modify mountpoints as needed to arrange to make the file syste

Re: [zfs-discuss] possible to override/inherit mountpoint on received snapshots?

2009-07-07 Thread Richard Elling
You need the zfs receive -u option. -- richard Andrew Daugherity wrote: I attempted to migrate data from one zfs pool to another, larger one (both pools are currently mounted on the same host) using the snapshot send/receive functionality. Of course, I could use something like rsync/cpio/tar

Re: [zfs-discuss] Problem with mounting ZFS from USB drive

2009-07-07 Thread Hua-Ying Ling
I'm no expert but I think you need to export a zfs volume before you remove it or it'll complain when you try to import it on another system. "zfs admin guide pg. 89" zfs export poolName You can do a zfs -f import to import it anyway. Hua-Ying On Tue, Jul 7, 2009 at 4:34 PM, Karl Dalen wrote:

Re: [zfs-discuss] [perf-discuss] help diagnosing system hang

2009-07-07 Thread Peter Eriksson
Interresting... I wonder what differs between your system and mine. With my dirt-simple stress-test: server1# zpool create X25E c1t15d0 server1# zfs set sharenfs=rw X25E server1# chmod a+w /X25E server2# cd /net/server1/X25E server2# gtar zxf /var/tmp/emacs-22.3.tar.gz and a fully patched X4242

Re: [zfs-discuss] Hanging receive

2009-07-07 Thread Ian Collins
Ian Collins wrote: Brent Jones wrote: On Fri, Jul 3, 2009 at 8:31 PM, Ian Collins wrote: Ian Collins wrote: I was doing an incremental send between pools, the receive side is locked up and no zfs/zpool commands work on that pool. The stacks look different from those reported in the ear

[zfs-discuss] possible to override/inherit mountpoint on received snapshots?

2009-07-07 Thread Andrew Daugherity
I attempted to migrate data from one zfs pool to another, larger one (both pools are currently mounted on the same host) using the snapshot send/receive functionality. Of course, I could use something like rsync/cpio/tar instead, but I'd have to first manually create all the target FSes, and se

[zfs-discuss] Problem with mounting ZFS from USB drive

2009-07-07 Thread Karl Dalen
I'm a new user of ZFS and I have an external USB drive which contains a ZFS pool with file system. It seems that it does not get auto mounted when I plug in the drive. I'm running osol-0811. How can I manually mount this drive? It has a pool named rpool on it. Is there any diagnostics commands tha

Re: [zfs-discuss] Disappearing snapshots

2009-07-07 Thread Tim Foster
On Mon, 2009-07-06 at 10:00 +0200, Juergen Nickelsen wrote: > DL Consulting writes: > Do not use the snapshots made for the time slider feature. These are > under control of the auto-snapshot service for exactly the time > slider and not for anything else. - or you could use the auto-snapshot:

[zfs-discuss] Adaptec SAS 2405 - anyone ?

2009-07-07 Thread roland
Hello, is anybody using this controller with opensolaris/snv ? http://www.adaptec.com/de-DE/products/Controllers/Hardware/sas/entry/SAS-2405/ does it run out of the box ? how does it perform with zfs? (especially when using it for zfs/nfs esx setup) the driver from adaptec is for solaris 10 u

Re: [zfs-discuss] Why is Solaris 10 ZFS performance so terrible?

2009-07-07 Thread Bob Friesenhahn
On Tue, 7 Jul 2009, Joerg Schilling wrote: Based on the prior discussions of using mmap() with ZFS and the way ZFS likes to work, my guess is that POSIX_FADV_NOREUSE does nothing at all and POSIX_FADV_DONTNEED probably does not work either. These are pretty straightforward to implement with UFS

Re: [zfs-discuss] Wrice cache for compressed file system

2009-07-07 Thread Gaëtan Lehmann
Le 7 juil. 09 à 15:54, Darren J Moffat a écrit : What compression algorithm are you using ? The default "on" value of lzjb or are you doing something like gzip-9 ? gzip-6. There is no speed problem with lzjb, but also not the same compression ratio :-) What build of OpenSolaris are you ru

Re: [zfs-discuss] Why is Solaris 10 ZFS performance so terrible?

2009-07-07 Thread Joerg Schilling
Bob Friesenhahn wrote: > On Tue, 7 Jul 2009, Joerg Schilling wrote: > > > > posix_fadvise seems to be _very_ new for Solaris and even though I am > > frequently reading/writing the POSIX standards mailing list, I was not > > aware of > > it. > > > > From my tests with star, I cannot see a signif

Re: [zfs-discuss] Why is Solaris 10 ZFS performance so terrible?

2009-07-07 Thread Bob Friesenhahn
On Tue, 7 Jul 2009, Joerg Schilling wrote: posix_fadvise seems to be _very_ new for Solaris and even though I am frequently reading/writing the POSIX standards mailing list, I was not aware of it. From my tests with star, I cannot see a significant performance increase but it may have a 3% effe

Re: [zfs-discuss] UCD-SNMP-MIB::dskAvail et.al. not supported on ZFS?

2009-07-07 Thread Richard Elling
Joerg Schilling wrote: Alexander Skwar wrote: Hi. I've got a fully patched Solaris 10 U7 Sparc system, on which I enabled SNMP disk monitoring by adding those lines to the /etc/sma/snmp/snmpd.conf configuration file: This is an OpenSolaris related list. Please repeat your tests on t

Re: [zfs-discuss] NFS, ZFS & ESX

2009-07-07 Thread Richard Elling
erik.ableson wrote: OK - I'm at my wit's end here as I've looked everywhere to find some means of tuning NFS performance with ESX into returning something acceptable using osol 2008.11. I've eliminated everything but the NFS portion of the equation and am looking for some pointers in the right

Re: [zfs-discuss] poor performance / lots of disk activity with low throughput, seems prefetch related

2009-07-07 Thread Richard Elling
Tom Bird wrote: Hi guys, I've been having trouble with my archival kit, in the performance department rather than data loss this time (phew!). At the point when I took these stats where was about 250 mbit of traffic outbound on an ixgb NIC on the thing, also about 100 mbit of new stuff inco

Re: [zfs-discuss] Why is Solaris 10 ZFS performance so terrible?

2009-07-07 Thread Gary Mills
On Mon, Jul 06, 2009 at 04:54:16PM +0100, Andrew Gabriel wrote: > Andre van Eyssen wrote: > >On Mon, 6 Jul 2009, Gary Mills wrote: > > > >>As for a business case, we just had an extended and catastrophic > >>performance degradation that was the result of two ZFS bugs. If we > >>have another one li

Re: [zfs-discuss] Why is Solaris 10 ZFS performance so terrible?

2009-07-07 Thread Joerg Schilling
James Andrewartha wrote: > Joerg Schilling wrote: > > I would be interested to see a open(2) flag that tells the system that I > > will > > read a file that I opened exactly once in native oder. This could tell the > > system to do read ahead and to later mark the pages as immediately > > reus

Re: [zfs-discuss] Wrice cache for compressed file system

2009-07-07 Thread Darren J Moffat
Gaëtan Lehmann wrote: Le 7 juil. 09 à 15:21, Darren J Moffat a écrit : Gaëtan Lehmann wrote: There will be two kinds of transfer protocol, once in production - CIFS and one specific to the application. But for a quick test, the test was made with scp. CIFS and scp are very different protoc

Re: [zfs-discuss] Zfs destroy taking inordinately long time...

2009-07-07 Thread Daniel Liebster
The destroy process must have hit point in the FS with a hundred thousand files. The destroy completed relatively quickly after passing that point. Please disregard this post. -- This message posted from opensolaris.org ___ zfs-discuss mailing list zfs

Re: [zfs-discuss] Wrice cache for compressed file system

2009-07-07 Thread Gaëtan Lehmann
Le 7 juil. 09 à 15:21, Darren J Moffat a écrit : Gaëtan Lehmann wrote: There will be two kinds of transfer protocol, once in production - CIFS and one specific to the application. But for a quick test, the test was made with scp. CIFS and scp are very different protocols with very differen

Re: [zfs-discuss] Wrice cache for compressed file system

2009-07-07 Thread Darren J Moffat
Gaëtan Lehmann wrote: There will be two kinds of transfer protocol, once in production - CIFS and one specific to the application. But for a quick test, the test was made with scp. CIFS and scp are very different protocols with very different performance characteristics. Also really importa

[zfs-discuss] Borrow disk from raidz1

2009-07-07 Thread Ares Drake
Hi all! I got a short question regarding data migration: I want to copy my data (~2TB) from an old machine to a new machine with a new raidz1 (6 disks x 1,5TB each ). Unfortunately this is not working properly via network due to various (driver) problems on the old machine. So my idea was: to

Re: [zfs-discuss] Wrice cache for compressed file system

2009-07-07 Thread Gaëtan Lehmann
Hi Darren, Le 7 juil. 09 à 13:41, Darren J Moffat a écrit : Gaëtan Lehmann wrote: I'd like to compress quite well compressable (~4x) data on a file server using ZFS compression, and still get good transfer speed. The users are transferring several GB of data (typically, 8-10 GB). The hos

[zfs-discuss] borrow disk from raidz?

2009-07-07 Thread Ares Drake
Hi all! I got a short question regarding data migration: I want to copy my data (~2TB) from an old machine to a new machine with a new raidz1 (6 disks x 1,5TB each ). Unfortunately this is not working properly via network due to various (driver) problems on the old machine. So my idea was: to b

Re: [zfs-discuss] UCD-SNMP-MIB::dskAvail et.al. not supported on ZFS?

2009-07-07 Thread Alexander Skwar
Hallo Jörg! On Tue, Jul 7, 2009 at 13:53, Joerg Schilling wrote: > Alexander Skwar wrote: > >> Hi. >> >> I've got a fully patched Solaris 10 U7 Sparc system, on which >> I enabled SNMP disk monitoring by adding those lines to the >> /etc/sma/snmp/snmpd.conf configuration file: > > This is an Open

Re: [zfs-discuss] UCD-SNMP-MIB::dskAvail et.al. not supported on ZFS?

2009-07-07 Thread Joerg Schilling
Alexander Skwar wrote: > Hi. > > I've got a fully patched Solaris 10 U7 Sparc system, on which > I enabled SNMP disk monitoring by adding those lines to the > /etc/sma/snmp/snmpd.conf configuration file: This is an OpenSolaris related list. Please repeat your tests on the current development pl

[zfs-discuss] UCD-SNMP-MIB::dskAvail et.al. not supported on ZFS?

2009-07-07 Thread Alexander Skwar
Hi. I've got a fully patched Solaris 10 U7 Sparc system, on which I enabled SNMP disk monitoring by adding those lines to the /etc/sma/snmp/snmpd.conf configuration file: disk / 5% disk /tmp 10% disk /data 5% That's supposed to mean, that I see <5% available on / to be critical, <10% on /tmp and

Re: [zfs-discuss] Wrice cache for compressed file system

2009-07-07 Thread Darren J Moffat
Gaëtan Lehmann wrote: I'd like to compress quite well compressable (~4x) data on a file server using ZFS compression, and still get good transfer speed. The users are transferring several GB of data (typically, 8-10 GB). The host is a X4150 with 16 GB of RAM. What protocol is being used for f

Re: [zfs-discuss] NFS, ZFS & ESX

2009-07-07 Thread Nicholas Lee
What is your NFS window size? 32kb * 120 * 7 should get you 25MB/s. Have considered getting a Intel X25-E?Going from 840 sync nfs iops to 3-5k+ iops is not overkill for SSD slog device. In fact probably cheaper to have one or two less vdevs and a single slog device. Nicholas On Tue, Jul 7,

Re: [zfs-discuss] surprisingly poor performance

2009-07-07 Thread James Andrewartha
James Lever wrote: > We also have a PERC 6/E w/512MB BBWC to test with or fall back to if we > go with a Linux solution. Have you tried putting the slog on this controller, either as an SSD or regular disk? It's supported by the mega_sas driver, x86 and amd64 only. -- James Andrewartha | Sysadmi

[zfs-discuss] NFS, ZFS & ESX

2009-07-07 Thread erik.ableson
OK - I'm at my wit's end here as I've looked everywhere to find some means of tuning NFS performance with ESX into returning something acceptable using osol 2008.11. I've eliminated everything but the NFS portion of the equation and am looking for some pointers in the right direction. Co

[zfs-discuss] poor performance / lots of disk activity with low throughput, seems prefetch related

2009-07-07 Thread Tom Bird
Hi guys, I've been having trouble with my archival kit, in the performance department rather than data loss this time (phew!). At the point when I took these stats where was about 250 mbit of traffic outbound on an ixgb NIC on the thing, also about 100 mbit of new stuff incoming. As you ca

Re: [zfs-discuss] Why is Solaris 10 ZFS performance so terrible?

2009-07-07 Thread James Andrewartha
Joerg Schilling wrote: > I would be interested to see a open(2) flag that tells the system that I will > read a file that I opened exactly once in native oder. This could tell the > system to do read ahead and to later mark the pages as immediately reusable. > This would make star even faster tha

Re: [zfs-discuss] recover data after zpool create

2009-07-07 Thread quasar
I've got the same problem. Did you find any solution? -- This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss