Re: [zfs-discuss] zfs
Hi Roman, from the provided data I suppose that you a running unpatched Solaris 10 Update 3. Since fault address is 0xc4 and in zio_create we manipulate mostly with zio_t structures, then 0xc4 most likely corresponds to io_child member of zio_t structure. If my assumption about Solaris update is correct, then corresponding piece of code is: zio_create+0x133: movl -0x4(%ebp),%edx zio_create+0x136: movl 0x1f0(%edx),%eax zio_create+0x13c: movl 0x1f4(%edx),%ecx zio_create+0x142: addl $0x1,%eax zio_create+0x145: adcl $0x0,%ecx zio_create+0x148: movl %eax,0x1f0(%edx) zio_create+0x14e: movl %ecx,0x1f4(%edx) zio_create+0x154: movl 0xc4(%edx),%eax zio_create+0x15a: movl %eax,0xcc(%ebx) zio_create+0x160: movl $0x0,0xc8(%ebx) zio_create+0x16a: movl 0xc4(%edx),%eax zio_create+0x170: testl %eax,%eax zio_create+0x172: je +0x8 zio_create+0x17a zio_create+0x174: movl %ebx,0xc8(%eax) zio_create+0x17a: movl -0x4(%ebp),%eax zio_create+0x17d: movl %ebx,0xc4(%eax) zio_create+0x183: movl 0x34(%eax),%eax zio_create+0x186: movl %eax,0x34(%ebx) zio_create+0x189: pushl %esi zio_create+0x18a: call +0x555fe07 mutex_exit Failure offset +0x17d corresponds to line 371 of zio.c http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/uts/common/fs/zfs/zio.c#371 Register %ebx contains address of newly allocates zio_t structure. At offset +0x17d we are trying to store %ebx into address %eax+0xc4, and since %eax is 0, we end up with page fault. We load %eax one instruction earlier from stack, so most probably something have overwritten stack very recently, since at offset 0x133 we loaded from that stack location into %edx (it was 0xe83edac0) and successfully dereferenced that address several times. Cheers, Victor Roman Chervotkin wrote: Hi. My system crashed today. System reboots without a problem and now everything looks as usual. By the way I used ztune.sh to tune parameters several days ago so the problem may be related to that script Is that a zfs issue or something different? Thanks, Roman --- -bash-3.00# more /var/adm/messages ... Apr 15 09:13:53 server3 unix: [ID 836849 kern.notice] Apr 15 09:13:53 server3 ^Mpanic[cpu1]/thread=dc28d600: Apr 15 09:13:53 server3 genunix: [ID 335743 kern.notice] BAD TRAP: type=e (#pf Page fault) rp=ecc53b9c addr=c4 occurred in module zfs due to a NULL pointer dereference Apr 15 09:13:53 server3 unix: [ID 10 kern.notice] Apr 15 09:13:53 server3 unix: [ID 839527 kern.notice] postgres: Apr 15 09:13:53 server3 unix: [ID 753105 kern.notice] #pf Page fault Apr 15 09:13:53 server3 unix: [ID 532287 kern.notice] Bad kernel fault at addr=0xc4 Apr 15 09:13:53 server3 unix: [ID 243837 kern.notice] pid=16204, pc=0xf7eda18c, sp=0x0, eflags=0x10282 Apr 15 09:13:53 server3 unix: [ID 211416 kern.notice] cr0: 80050033pg,wp,ne,et,mp,pe cr4: 6d8xmme,fxsr,pge,mce,pse,de Apr 15 09:13:53 server3 unix: [ID 936844 kern.notice] cr2: c4 cr3: 66e8e000 Apr 15 09:13:53 server3 unix: [ID 537610 kern.notice]gs: 1b0 fs: e339 es: ecc50160 ds: f7ed0160 Apr 15 09:13:53 server3 unix: [ID 537610 kern.notice] edi:d esi: e83edcbc ebp: ecc53bfc esp: ecc53bd4 Apr 15 09:13:53 server3 unix: [ID 537610 kern.notice] ebx: e3390cc0 edx: e83edac0 ecx:0 eax:0 Apr 15 09:13:53 server3 unix: [ID 537610 kern.notice] trp:e err: 2 eip: f7eda18c cs: 158 Apr 15 09:13:53 server3 unix: [ID 717149 kern.notice] efl:10282 usp: 0 ss: eb6f46d8 Apr 15 09:13:53 server3 unix: [ID 10 kern.notice] Apr 15 09:13:53 server3 genunix: [ID 353471 kern.notice] ecc53afc unix:die+a7 (e, ecc53b9c, c4, 1) Apr 15 09:13:53 server3 genunix: [ID 353471 kern.notice] ecc53b88 unix:trap+103f (ecc53b9c, c4, 1) Apr 15 09:13:53 server3 genunix: [ID 353471 kern.notice] ecc53b9c unix:_cmntrap+9a (1b0, e339, ecc5) Apr 15 09:13:53 server3 genunix: [ID 353471 kern.notice] ecc53bfc zfs:zio_create+17d (e83edac0, d8969900,) Apr 15 09:13:53 server3 genunix: [ID 353471 kern.notice] ecc53c44 zfs:zio_vdev_child_io+67 (e83edac0, e7d27d8c,) Apr 15 09:13:53 server3 genunix: [ID 353471 kern.notice] ecc53c98 zfs:vdev_mirror_io_start+14d (e83edac0, ecc53cc8,) Apr 15 09:13:53 server3 genunix: [ID 353471 kern.notice] ecc53ca4 zfs:vdev_io_start+15 (e83edac0) Apr 15 09:13:53 server3 genunix: [ID 353471 kern.notice] ecc53cc8 zfs:zio_vdev_io_start+13f (e83edac0) Apr 15 09:13:53 server3 genunix: [ID 353471 kern.notice] ecc53cd8 zfs:zfsctl_ops_root+2044461b (e83edac0, ecc53d04,) Apr 15 09:13:53 server3 genunix: [ID 353471 kern.notice] ecc53ce4 zfs:zio_nowait+b (e83edac0) Apr 15 09:13:53 server3 genunix: [ID 353471 kern.notice] ecc53d04
[zfs-discuss] snapshot features
hi all , when doing several zfs snapshot of a given fs, there are dependencies between snapshots that complexify the management of snapshots is there a plan to easy thes dependencies, so we can reach snapshot functionalities that are offered in other products suchs as Compellent (http://www.compellent.com/products/software/continuous_snapshots.aspx) Compellent software allows to set **retention periods** for different snapshots and will manage their migration or deletion automatically thanks s. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS and Linux
Nicolas Williams [EMAIL PROTECTED] wrote: Sigh. We have devolved. Every thread on OpenSolaris discuss lists seems to devolve into a license discussion. It is funny to see that in our case, the tecnical problems (those caused by the fact that linux implements a different VFS interface layer) are creating much bigger problem than the license issue does. I have seen mailing list posts (I'd have to search again) that indicate [that some believe] that even dynamic linking via dlopen() qualifies as making a derivative. There is no single place in the GPL that mentions the term linking. For this reason, the GPL FAQ from the FSF is wring as it is based on the term linking. There is no difference whether you link statically or dynamically. Whether using GPLd code from a non-GPLd program creates a derived work thus cannot depend on whether you link agaist it or not. If a GPLd program however uses a non-GPLd library, this is definitely not a problem or every GPLd program linked against the libc from HP-UX would be a problem. If true that would mean that one could not distribute an OpenSolaris distribution containing a GPLed PAM module. Or perhaps, because in that case the header files needed to make the linking possible are not GPLed the linking-makes-derivatives argument would not apply. If the GPLd PAM module just implements a well known plug in interface, a program that uses this odule cannot be a derivate of the GPLd code. Jörg -- EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED](uni) [EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS and Linux
Paul Fisher [EMAIL PROTECTED] wrote: Is there any reason that the CDDL dictates, or that Sun would object, to zfs being made available as an independently distributed Linux kernel module? In other words, if I made an Nvidia-like distribution available, would that be OK from the OpenSolaris side? The way I understand the fact/way that Sun did openSource Solaris is that there are no objections from Sun's side against using OpenSolaris code inside other projects. Jörg -- EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED](uni) [EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Testing of UFS, VxFS and ZFS
Did you measure CPU utilization by any chance during the tests? Its T2000 and CPU cores are quite slow on this box hence might be a bottleneck. just a guess. On Mon, 2007-04-16 at 13:10 -0400, Tony Galway wrote: I had previously undertaken a benchmark that pits “out of box” performance of UFS via SVM, VxFS and ZFS but was waylaid due to some outstanding availability issues in ZFS. These have been taken care of, and I am once again undertaking this challenge on behalf of my customer. The idea behind this benchmark is to show a. How ZFS might displace the current commercial volume and file system management applications being used. b. The learning curve of moving from current volume management products to ZFS. c. Performance differences across the different volume management products. VDBench is the test bed of choice as this has been accepted by the customer as a telling and accurate indicator of performance. The last time I attempted this test it had been suggested that VDBench is not appropriate to testing ZFS, I cannot see that being a problem, VDBench is a tool – if it highlights performance problems, then I would think it is a very effective tool so that we might better be able to fix those deficiencies. Now, to the heart of my problem! The test hardware is a T2000 connected to a 12 disk SE3510 (presenting as JBOD) through a brocade switch, and I am using Solaris 10 11/06. For Veritas, I am using Storage Foundation Suite 5.0. The systems were jumpstarted to the same configuration before testing a different volume management software to ensure there were no artifacts remaining from any previous test. I present my vdbench definition below for your information: sd=FS,lun=/pool/TESTFILE,size=10g,threads=8 wd=DWR,sd=FS,rdpct=100,seekpct=80 wd=ETL,sd=FS,rdpct=0, seekpct=80 wd=OLT,sd=FS,rdpct=70, seekpct=80 rd=R1-DWR,wd=DWR,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8k,16k,32k,64k,128k) rd=R1-ETL,wd=ETL,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8k,16k,32k,64k,128k) rd=R1-OLT,wd=OLT,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8k,16k,32k,64k,128k) rd=R2-DWR,wd=DWR,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8k,16k,32k,64k,128k) rd=R2-ETL,wd=ETL,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8k,16k,32k,64k,128k) rd=R2-OLT,wd=OLT,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8k,16k,32k,64k,128k) rd=R3-DWR,wd=DWR,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8k,16k,32k,64k,128k) rd=R3-ETL,wd=ETL,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8k,16k,32k,64k,128k) rd=R3-OLT,wd=OLT,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8k,16k,32k,64k,128k) As you can see, it is fairly straight forward and I take the average of the three runs in each of ETL, OLT and DWR workloads. As an aside, I am also performing this test for various file system block sizes as applicable as well. I then ran this workload against a Raid-5 LUN created and mounted in each of the different file system types. Please note that one of the test criteria is that the associated volume management software create the Raid-5 LUN, not the disk subsystem. 1. UFS via SVM # metainit d20 –r d1 … d8 # newfs /dev/md/dsk/d20 # mount /dev/md/dsk/d20 /pool 2. ZFS # zfs create pool raidz d1 … d8 3. VxFS – Veritas SF5.0 # vxdisk init SUN35100_0 …. SUN35100_7 # vxdg init testdg SUN35100_0 … # vxassist –g testdg make pool 418283m layout=raid5 Now to my problem – Performance! Given the test as defined above, VxFS absolutely blows the doors off of both UFS and ZFS during write operations. For example, during a single test on an 8k file system block, I have the following average IO Rates: ETL OLTP DWR UFS 390.00 1298.44 23173.60 VxFS 15323.10 27329.04 22889.91 ZFS 2122.23 7299.36 22940.63 If you look at these numbers percentage wise, with VxFS being set to 100%, then UFS run’s at 2.5% the speed, and ZFS at 13.8% the speed, for OLTP UFS is 4.8% and ZFS 26.7%, however in DWR where there are 100% reads, no writing, performance is similar with UFS at 101.2% and ZFS at 100.2% the speed of VxFS. cid:image002.png@01C78027.99B515D0 Given this performance problems, then quite obviously VxFS quite rightly deserves to be the file system of choice, even with a cost premium. If anyone has any insight into why I am seeing, consistently, these types of very disappointing numbers I would very much appreciate your comments. The numbers are very disturbing as it is indicating that write
Re: [zfs-discuss] Testing of UFS, VxFS and ZFS
On April 16, 2007 1:10:41 PM -0400 Tony Galway [EMAIL PROTECTED] wrote: I had previously undertaken a benchmark that pits out of box performance ... The test hardware is a T2000 connected to a 12 disk SE3510 (presenting as ... Now to my problem - Performance! Given the test as defined above, VxFS absolutely blows the doors off of both UFS and ZFS during write operations. For example, during a single test on an 8k file system block, I have the following average IO Rates: out of the box performance of zfs on T2000 hardware might suffer. http://blogs.sun.com/realneel/entry/zfs_and_databases is the only link I could find, but there is another article somewhere about tuning for t2000, related to PCI on the t2000, ie it is t2000-specific. -frank ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
RE: [zfs-discuss] Testing of UFS, VxFS and ZFS
The volume is 7+1. I have created the volume using both the default (DRL) as well as 'nolog' to turn it off, both with similar performance. On the advice of Henk, after he had looked over my data, he is notice that the veritas test seems to be almost entirely using file system cache. I will retest with a much larger file to defeat this cache (I do not want to modify my mount options). If this then shows similar performance (I will also retest with ZFS with the same file size) then the question will probably have more to do with how ZFS handles file system caching. -Tony -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: April 16, 2007 2:16 PM To: [EMAIL PROTECTED] Subject: Re: [zfs-discuss] Testing of UFS, VxFS and ZFS Is the VxVM volume 8-wide? It is not clear from your creation commands. -- richard Tony Galway wrote: I had previously undertaken a benchmark that pits out of box performance of UFS via SVM, VxFS and ZFS but was waylaid due to some outstanding availability issues in ZFS. These have been taken care of, and I am once again undertaking this challenge on behalf of my customer. The idea behind this benchmark is to show a. How ZFS might displace the current commercial volume and file system management applications being used. b. The learning curve of moving from current volume management products to ZFS. c. Performance differences across the different volume management products. VDBench is the test bed of choice as this has been accepted by the customer as a telling and accurate indicator of performance. The last time I attempted this test it had been suggested that VDBench is not appropriate to testing ZFS, I cannot see that being a problem, VDBench is a tool - if it highlights performance problems, then I would think it is a very effective tool so that we might better be able to fix those deficiencies. Now, to the heart of my problem! The test hardware is a T2000 connected to a 12 disk SE3510 (presenting as JBOD) through a brocade switch, and I am using Solaris 10 11/06. For Veritas, I am using Storage Foundation Suite 5.0. The systems were jumpstarted to the same configuration before testing a different volume management software to ensure there were no artifacts remaining from any previous test. I present my vdbench definition below for your information: sd=FS,lun=/pool/TESTFILE,size=10g,threads=8 wd=DWR,sd=FS,rdpct=100,seekpct=80 wd=ETL,sd=FS,rdpct=0, seekpct=80 wd=OLT,sd=FS,rdpct=70, seekpct=80 rd=R1-DWR,wd=DWR,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8 k,16k,32k,64k,128k) rd=R1-ETL,wd=ETL,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8 k,16k,32k,64k,128k) rd=R1-OLT,wd=OLT,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8 k,16k,32k,64k,128k) rd=R2-DWR,wd=DWR,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8 k,16k,32k,64k,128k) rd=R2-ETL,wd=ETL,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8 k,16k,32k,64k,128k) rd=R2-OLT,wd=OLT,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8 k,16k,32k,64k,128k) rd=R3-DWR,wd=DWR,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8 k,16k,32k,64k,128k) rd=R3-ETL,wd=ETL,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8 k,16k,32k,64k,128k) rd=R3-OLT,wd=OLT,iorate=max,elapsed=1800,interval=30,forxfersize=(1k,2k,4k,8 k,16k,32k,64k,128k) As you can see, it is fairly straight forward and I take the average of the three runs in each of ETL, OLT and DWR workloads. As an aside, I am also performing this test for various file system block sizes as applicable as well. I then ran this workload against a Raid-5 LUN created and mounted in each of the different file system types. Please note that one of the test criteria is that the associated volume management software create the Raid-5 LUN, not the disk subsystem. 1. UFS via SVM # metainit d20 -r d1 . d8 # newfs /dev/md/dsk/d20 # mount /dev/md/dsk/d20 /pool 2. ZFS # zfs create pool raidz d1 . d8 3. VxFS - Veritas SF5.0 # vxdisk init SUN35100_0 .. SUN35100_7 # vxdg init testdg SUN35100_0 . # vxassist -g testdg make pool 418283m layout=raid5 Now to my problem - Performance! Given the test as defined above, VxFS absolutely blows the doors off of both UFS and ZFS during write operations. For example, during a single test on an 8k file system block, I have the following average IO Rates: *ETL * *OLTP * *DWR * *UFS * 390.00 1298.44 23173.60 *VxFS * 15323.10 27329.04 22889.91 *ZFS * 2122.23 7299.36 22940.63 If you look at these
[zfs-discuss] Re: Testing of UFS, VxFS and ZFS
Why are you using software-based RAID 5/RAIDZ for the tests? I didn't think this was a common setup in cases where file system performance was the primary consideration. 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 question as to sizes
[EMAIL PROTECTED] wrote on 04/16/2007 04:57:43 PM: one pool is mirror on 300gb dirives and the other is raidz1 on 7 x 143gb drives. I did make clone of my zfs file systems with their snaps and something is not right, sizes do not match... anyway here is what I have: [17:50:32] [EMAIL PROTECTED]: /root zfs list NAME USED AVAIL REFER MOUNTPOINT mypool 272G 1.95G 24.5K /mypool mypool/d 272G 1.95G 143G /d/d2 mypool/[EMAIL PROTECTED] 3.72G - 123G - mypool/[EMAIL PROTECTED] 22.3G - 156G - mypool/[EMAIL PROTECTED] 23.3G - 161G - mypool/[EMAIL PROTECTED] 16.1G - 172G - mypool/[EMAIL PROTECTED] 13.8G - 168G - mypool/[EMAIL PROTECTED] 15.7G - 168G - mypool/[EMAIL PROTECTED]192M - 143G - mypool/[EMAIL PROTECTED]189M - 143G - mypool/[EMAIL PROTECTED]200M - 143G - mypool/[EMAIL PROTECTED] 3.93M - 143G - mypool2463G 474G52K /mypool2 mypool2/d 318G 474G 168G /mypool2/d mypool2/[EMAIL PROTECTED] 4.40G - 145G - mypool2/[EMAIL PROTECTED] 26.1G - 184G - mypool2/[EMAIL PROTECTED] 27.3G - 189G - mypool2/[EMAIL PROTECTED] 18.7G - 202G - mypool2/[EMAIL PROTECTED] 16.1G - 197G - mypool2/[EMAIL PROTECTED]18.2G - 198G - mypool2/d3 145G 474G 145G legacy see: mypool/d 272G 1.95G 143G /d/d2 mypool2/d 318G 474G 168G /mypool2/d they are the same copies but their sizes do differ quite a bit, original is 272G and the clone which I duplicated by zfs send/receive is 318G. Then all the other snaps also do differ. Why is that difference? Could someone explain how does it work? No, snapshot space usage reporting is beyond brain-dead at this point, shared data between snaps is hidden from the reporting and there is no way (short of deleting snapshots) to see how much space they are holding hostage. The ZFS team has said that they are working on providing more detail -- I have not seen anything yet. Why it was considered a valid data column in its current state is anyone's guess -- in one of my servers I have over 7 tb unaccounted for in the zfs listing because of this issue. -Wade ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] zfs send/receive question
[18:19:00] [EMAIL PROTECTED]: /root zfs send -i mypool/[EMAIL PROTECTED] mypool/[EMAIL PROTECTED] | zfs receive -F mypool2/[EMAIL PROTECTED] invalid option 'F' usage: receive [-vn] filesystem|volume|snapshot receive [-vn] -d filesystem For the property list, run: zfs set|get It does not seem to work unless I am doing it incorectly. Chris On Tue, 17 Apr 2007, Nicholas Lee wrote: On 4/17/07, Krzys [EMAIL PROTECTED] wrote: and when I did try to run that last command I got the following error: [16:26:00] [EMAIL PROTECTED]: /root zfs send -i mypool/[EMAIL PROTECTED] mypool/[EMAIL PROTECTED] | zfs receive mypool2/[EMAIL PROTECTED] cannot receive: destination has been modified since most recent snapshot is there any way to do such replication by zfs send/receive and avoind such error message? Is there any way to force file system not to be mounted? Is there any way to make it maybe read only parition and then when its needed maybe make it live or whaverer? Check the -F option to zfs receive. This automatically rolls back the target. Nicholas ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] ZFS, Multiple Machines and NFS
Hi! I am very new to ZFS (never installed it), and I have a small question. Is it possible with ZFS to merge multiple machines with NFS into one ZFS filesystem so they look like one storage device? As I'm typing this I feel like a fool, but I'll ask anyway. :-) Thanks! -=//-\drian Thompson=- ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] zfs question as to sizes
On Mon, Apr 16, 2007 at 05:13:37PM -0500, [EMAIL PROTECTED] wrote: Why it was considered a valid data column in its current state is anyone's guess. This column is precise and valid. It represents the amount of space uniquely referenced by the snapshot, and therefore the amount of space that would be freed were it to be deleted. The shared space between snapshots, besides being difficult to calculate, is nearly impossible to enumerate in anything beyond the most trivial setups. For example, with just snapshots 'a b c d e', you can have space shared by the following combinations: a b a b c a b c d a b c d e b c b c d b c d e c d c d e d e Not to mention the space shared with the active filesystem. With dozens of snapshots, you're talking about hundreds or thousands of combinations. It's certainly possible to calculate the space used various snapshot intersections, but presenting it is another matter. Perhaps you could describe how you would like this information to be presented in zfs(1M). - Eric -- Eric Schrock, Solaris Kernel Development http://blogs.sun.com/eschrock ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] zfs send/receive question
Chris, Looks like you're not running a Solaris release that contains the zfs receive -F option. This option is in current Solaris community release, build 48. http://docs.sun.com/app/docs/doc/817-2271/6mhupg6f1?a=view#gdsup Otherwise, you'll have to wait until an upcoming Solaris 10 release. Cindy Krzys wrote: [18:19:00] [EMAIL PROTECTED]: /root zfs send -i mypool/[EMAIL PROTECTED] mypool/[EMAIL PROTECTED] | zfs receive -F mypool2/[EMAIL PROTECTED] invalid option 'F' usage: receive [-vn] filesystem|volume|snapshot receive [-vn] -d filesystem For the property list, run: zfs set|get It does not seem to work unless I am doing it incorectly. Chris On Tue, 17 Apr 2007, Nicholas Lee wrote: On 4/17/07, Krzys [EMAIL PROTECTED] wrote: and when I did try to run that last command I got the following error: [16:26:00] [EMAIL PROTECTED]: /root zfs send -i mypool/[EMAIL PROTECTED] mypool/[EMAIL PROTECTED] | zfs receive mypool2/[EMAIL PROTECTED] cannot receive: destination has been modified since most recent snapshot is there any way to do such replication by zfs send/receive and avoind such error message? Is there any way to force file system not to be mounted? Is there any way to make it maybe read only parition and then when its needed maybe make it live or whaverer? Check the -F option to zfs receive. This automatically rolls back the target. Nicholas ___ 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] crashed remote system trying to do zfs send / receive
Hello Krzys, Sunday, April 15, 2007, 4:53:43 AM, you wrote: K Strange thing, I did try to do zfs send/receive using zfs. K On the from host I did the following: K bash-3.00# zfs send mypool/zones/[EMAIL PROTECTED] | ssh 10.0.2.79 zfs receive K mypool/zones/[EMAIL PROTECTED] K Password: K ^CKilled by signal 2. K 1 or 2 minutes later I did break this command and I wanted to time it so I did K change command and reissued it. K bash-3.00# time zfs send mypool/zones/[EMAIL PROTECTED] | ssh 10.0.2.79 zfs K receive mypool/zones/[EMAIL PROTECTED] K Password: K ^CKilled by signal 2. K real0m7.346s K user0m0.220s K sys 0m0.036s K bash-3.00# K Right after this I got on remote server kernel panic and here is the output: K [22:35:30] @zglobix1: /root K panic[cpu1]/thread=30001334380: dangling dbufs (dn=6000a13eba0, K dbuf=60007d927e8) K 02a1004f9030 zfs:dnode_evict_dbufs+19c (6000a13eba0, 1, 7b64e800, K 6000a9e34b0, 1, 2a1004f90e8) K%l0-3: 06000a13edb0 06000a13edb8 K%l4-7: 02a1004f90e8 060007be3910 0003 0001 K 02a1004f9230 zfs:dmu_objset_evict_dbufs+d8 (21, 0, 0, 7b648400, 6000a9e32c0, K 6000a9e32c0) K%l0-3: 060002738f89 000f 060007fef8f0 06000a9e34a0 K%l4-7: 06000a13eba0 06000a9e3398 0001 7b6485e7 K 02a1004f92e0 zfs:dmu_objset_evict+b4 (60007138900, 6000a9e32c0, 180e580, K 7b60a800, 7b64e400, 7b64e400) K%l0-3: 0180c000 0005 K%l4-7: 060007b9f600 7b60a800 7b64e400 K 02a1004f93a0 zfs:dsl_dataset_evict+34 (60007138900, 7b60af7c, 18364c0, K 60001ac90c0, 6000a9e32c0, 60007b9f600) K%l0-3: 02a10001fcc0 02a10001fcc0 K%l4-7: 030b1b80 01ac8d1a 018a8400 K 02a1004f9450 zfs:dbuf_evict_user+48 (60007138908, 60007b9f600, 60008666cd0, K 0, 0, 60008666be8) K%l0-3: 060007138900 0013 K%l4-7: 03000107c000 018ade70 0bc0 7b612fa4 K 02a1004f9500 zfs:zfsctl_ops_root+b184ac4 (60008666c40, 60008666be8, K 70478000, 3, 3, 0) K%l0-3: 060001ac90c0 000f 0600071389a0 K%l4-7: 0001 70478018 K 02a1004f95b0 zfs:dmu_recvbackup+8e8 (60006f32d00, 60006f32fd8, 60006f32e30, K 1, 60006ad5fa8, 0) K%l0-3: 060006f32d15 060007138900 7b607c00 7b648000 K%l4-7: 0040 0354 0001 0138 K 02a1004f9780 zfs:zfs_ioc_recvbackup+38 (60006f32000, 0, 0, 0, 9, 0) K%l0-3: 0004 006d K%l4-7: 060006f3200c 0073 K 02a1004f9830 zfs:zfsdev_ioctl+160 (70478c00, 5d, ffbfebc8, 1f, 7c, 1000) K%l0-3: 060006f32000 007c K%l4-7: 7b63b688 70479248 02e8 70478f60 K 02a1004f98e0 genunix:fop_ioctl+20 (60006c8fd00, 5a1f, ffbfebc8, 13, K 600067d34b0, 12066d4) K%l0-3: 0600064da200 0600064da200 0003 060006cc6fd8 K%l4-7: ff342036 ff345c7c 018a9400 K 02a1004f9990 genunix:ioctl+184 (3, 60006c03688, ffbfebc8, 6500, ff00, K 5a1f) K%l0-3: 0004 c1ac K%l4-7: 0001 K syncing file systems... 2 1 done K dumping to /dev/dsk/c1t1d0s1, offset 3436642304, content: kernel K 94% done K SC Alert: Failed to send email alert to the primary mailserver. K SC Alert: Failed to send email alert for recent event. K 100% done: 71483 pages dumped, compression ratio 5.34, dump succeeded K rebooting... K SC Alert: Host System has Reset K Probing system devices K Probing memory K Probing I/O buses K Sun Fire V240, No Keyboard K Copyright 1998-2004 Sun Microsystems, Inc. All rights reserved. K OpenBoot 4.16.2, 16384 MB memory installed, Serial #63395381. K Ethernet address 0:3:ba:c7:56:35, Host ID: 83c75635. K Initializing 2048MB of memory at addr10 \ K SC Alert: Failed to send email alert for recent event. K Rebooting with command: boot K Boot device: disk1 File and args: K SunOS Release 5.10 Version Generic_118833-36 64-bit K Copyright 1983-2006 Sun Microsystems, Inc. All rights reserved. K Use is subject to license terms. K / K SC Alert: Failed to send email alert for recent event. K Hardware watchdog enabled K Hostname: zglobix1 K checking ufs filesystems K /dev/rdsk/c1t1d0s7: is logging. K Failed to send email alert for recent event. K SC Alert: Failed to send email alert for recent event. K zglobix1 console login: K Any
Re: [zfs-discuss] ZFS, Multiple Machines and NFS
Adrian, you can take a look at pNFS: http://opensolaris.org/os/community/os_user_groups/frosug/pNFS/FROSUG-pNFS.pdf Project homepage: http://opensolaris.org/os/project/nfsv41/ Rayson On 4/16/07, Jason A. Hoffman [EMAIL PROTECTED] wrote: On Apr 16, 2007, at 3:24 PM, Adrian Thompson wrote: Hi! I am very new to ZFS (never installed it), and I have a small question. Is it possible with ZFS to merge multiple machines with NFS into one ZFS filesystem so they look like one storage device? As I'm typing this I feel like a fool, but I'll ask anyway. :-) Thanks! -=//-\drian Thompson=- Not a foolish question at all, but the answer is no. ZFS works with block devices. Regards, Jason ___ 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 send/receive question
Ah, ok, not a problem, do you know Cindy when next Solaris Update is going to be released by SUN? Yes, I am running U3 at this moment. Regards, Chris On Mon, 16 Apr 2007, [EMAIL PROTECTED] wrote: Chris, Looks like you're not running a Solaris release that contains the zfs receive -F option. This option is in current Solaris community release, build 48. http://docs.sun.com/app/docs/doc/817-2271/6mhupg6f1?a=view#gdsup Otherwise, you'll have to wait until an upcoming Solaris 10 release. Cindy Krzys wrote: [18:19:00] [EMAIL PROTECTED]: /root zfs send -i mypool/[EMAIL PROTECTED] mypool/[EMAIL PROTECTED] | zfs receive -F mypool2/[EMAIL PROTECTED] invalid option 'F' usage: receive [-vn] filesystem|volume|snapshot receive [-vn] -d filesystem For the property list, run: zfs set|get It does not seem to work unless I am doing it incorectly. Chris On Tue, 17 Apr 2007, Nicholas Lee wrote: On 4/17/07, Krzys [EMAIL PROTECTED] wrote: and when I did try to run that last command I got the following error: [16:26:00] [EMAIL PROTECTED]: /root zfs send -i mypool/[EMAIL PROTECTED] mypool/[EMAIL PROTECTED] | zfs receive mypool2/[EMAIL PROTECTED] cannot receive: destination has been modified since most recent snapshot is there any way to do such replication by zfs send/receive and avoind such error message? Is there any way to force file system not to be mounted? Is there any way to make it maybe read only parition and then when its needed maybe make it live or whaverer? Check the -F option to zfs receive. This automatically rolls back the target. Nicholas ___ 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 !DSPAM:122,4623fa8a1809423226276! ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] crashed remote system trying to do zfs send / receive
Ah, perfect then... Thank you so much for letting me know... Regards, Chris On Tue, 17 Apr 2007, Robert Milkowski wrote: Hello Krzys, Sunday, April 15, 2007, 4:53:43 AM, you wrote: K Strange thing, I did try to do zfs send/receive using zfs. K On the from host I did the following: K bash-3.00# zfs send mypool/zones/[EMAIL PROTECTED] | ssh 10.0.2.79 zfs receive K mypool/zones/[EMAIL PROTECTED] K Password: K ^CKilled by signal 2. K 1 or 2 minutes later I did break this command and I wanted to time it so I did K change command and reissued it. K bash-3.00# time zfs send mypool/zones/[EMAIL PROTECTED] | ssh 10.0.2.79 zfs K receive mypool/zones/[EMAIL PROTECTED] K Password: K ^CKilled by signal 2. K real0m7.346s K user0m0.220s K sys 0m0.036s K bash-3.00# K Right after this I got on remote server kernel panic and here is the output: K [22:35:30] @zglobix1: /root K panic[cpu1]/thread=30001334380: dangling dbufs (dn=6000a13eba0, K dbuf=60007d927e8) K 02a1004f9030 zfs:dnode_evict_dbufs+19c (6000a13eba0, 1, 7b64e800, K 6000a9e34b0, 1, 2a1004f90e8) K%l0-3: 06000a13edb0 06000a13edb8 K%l4-7: 02a1004f90e8 060007be3910 0003 0001 K 02a1004f9230 zfs:dmu_objset_evict_dbufs+d8 (21, 0, 0, 7b648400, 6000a9e32c0, K 6000a9e32c0) K%l0-3: 060002738f89 000f 060007fef8f0 06000a9e34a0 K%l4-7: 06000a13eba0 06000a9e3398 0001 7b6485e7 K 02a1004f92e0 zfs:dmu_objset_evict+b4 (60007138900, 6000a9e32c0, 180e580, K 7b60a800, 7b64e400, 7b64e400) K%l0-3: 0180c000 0005 K%l4-7: 060007b9f600 7b60a800 7b64e400 K 02a1004f93a0 zfs:dsl_dataset_evict+34 (60007138900, 7b60af7c, 18364c0, K 60001ac90c0, 6000a9e32c0, 60007b9f600) K%l0-3: 02a10001fcc0 02a10001fcc0 K%l4-7: 030b1b80 01ac8d1a 018a8400 K 02a1004f9450 zfs:dbuf_evict_user+48 (60007138908, 60007b9f600, 60008666cd0, K 0, 0, 60008666be8) K%l0-3: 060007138900 0013 K%l4-7: 03000107c000 018ade70 0bc0 7b612fa4 K 02a1004f9500 zfs:zfsctl_ops_root+b184ac4 (60008666c40, 60008666be8, K 70478000, 3, 3, 0) K%l0-3: 060001ac90c0 000f 0600071389a0 K%l4-7: 0001 70478018 K 02a1004f95b0 zfs:dmu_recvbackup+8e8 (60006f32d00, 60006f32fd8, 60006f32e30, K 1, 60006ad5fa8, 0) K%l0-3: 060006f32d15 060007138900 7b607c00 7b648000 K%l4-7: 0040 0354 0001 0138 K 02a1004f9780 zfs:zfs_ioc_recvbackup+38 (60006f32000, 0, 0, 0, 9, 0) K%l0-3: 0004 006d K%l4-7: 060006f3200c 0073 K 02a1004f9830 zfs:zfsdev_ioctl+160 (70478c00, 5d, ffbfebc8, 1f, 7c, 1000) K%l0-3: 060006f32000 007c K%l4-7: 7b63b688 70479248 02e8 70478f60 K 02a1004f98e0 genunix:fop_ioctl+20 (60006c8fd00, 5a1f, ffbfebc8, 13, K 600067d34b0, 12066d4) K%l0-3: 0600064da200 0600064da200 0003 060006cc6fd8 K%l4-7: ff342036 ff345c7c 018a9400 K 02a1004f9990 genunix:ioctl+184 (3, 60006c03688, ffbfebc8, 6500, ff00, K 5a1f) K%l0-3: 0004 c1ac K%l4-7: 0001 K syncing file systems... 2 1 done K dumping to /dev/dsk/c1t1d0s1, offset 3436642304, content: kernel K 94% done K SC Alert: Failed to send email alert to the primary mailserver. K SC Alert: Failed to send email alert for recent event. K 100% done: 71483 pages dumped, compression ratio 5.34, dump succeeded K rebooting... K SC Alert: Host System has Reset K Probing system devices K Probing memory K Probing I/O buses K Sun Fire V240, No Keyboard K Copyright 1998-2004 Sun Microsystems, Inc. All rights reserved. K OpenBoot 4.16.2, 16384 MB memory installed, Serial #63395381. K Ethernet address 0:3:ba:c7:56:35, Host ID: 83c75635. K Initializing 2048MB of memory at addr10 \ K SC Alert: Failed to send email alert for recent event. K Rebooting with command: boot K Boot device: disk1 File and args: K SunOS Release 5.10 Version Generic_118833-36 64-bit K Copyright 1983-2006 Sun Microsystems, Inc. All rights reserved. K Use is subject to license terms. K / K SC Alert: Failed to send email alert for recent event. K Hardware watchdog enabled K Hostname: zglobix1 K checking ufs filesystems K /dev/rdsk/c1t1d0s7: is logging. K Failed to
Re: [zfs-discuss] zfs send/receive question
On 16/04/07, Krzys [EMAIL PROTECTED] wrote: Ah, ok, not a problem, do you know Cindy when next Solaris Update is going to be released by SUN? Yes, I am running U3 at this moment. Summer is what I last read (July?). -- Less is only more where more is no good. --Frank Lloyd Wright Shawn Walker, Software and Systems Analyst [EMAIL PROTECTED] - http://binarycrusader.blogspot.com/ ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] How to bind the oracle 9i data file to zfs volumes
Simon wrote: So,does mean this is oracle bug ? Or it's impossible(or inappropriate) to use ZFS/SVM volumes to create oracle data file,instead,should use zfs or ufs filesystem to do this. Oracle can use SVM volumes to hold its data. Unless I am mistaken, it should be able to use zvols as well. However, googling for 'zvol + Oracle' did not get me anything useful. Perhaps it is not a configuration that is very popular. ;) My $ 0.02. -Manoj ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] snapshot features
Frank Cusack wrote: On April 16, 2007 10:24:04 AM +0200 Selim Daoud [EMAIL PROTECTED] wrote: hi all , when doing several zfs snapshot of a given fs, there are dependencies between snapshots that complexify the management of snapshots is there a plan to easy thes dependencies, so we can reach snapshot functionalities that are offered in other products suchs as Compellent (http://www.compellent.com/products/software/continuous_snapshots.aspx) Compellent software allows to set **retention periods** for different snapshots and will manage their migration or deletion automatically retention period is pretty easily managed via cron Yeah but cron isn't easily managed by anything. :-P ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Testing of UFS, VxFS and ZFS
Tony Galway wrote: I had previously undertaken a benchmark that pits “out of box” performance of UFS via SVM, VxFS and ZFS but was waylaid due to some outstanding availability issues in ZFS. These have been taken care of, and I am once again undertaking this challenge on behalf of my customer. The idea behind this benchmark is to show a. How ZFS might displace the current commercial volume and file system management applications being used. b. The learning curve of moving from current volume management products to ZFS. c. Performance differences across the different volume management products. VDBench is the test bed of choice as this has been accepted by the customer as a telling and accurate indicator of performance. The last time I attempted this test it had been suggested that VDBench is not appropriate to testing ZFS, I cannot see that being a problem, VDBench is a tool – if it highlights performance problems, then I would think it is a very effective tool so that we might better be able to fix those deficiencies. First, VDBench is a Sun internal and partner only tool so you might not get much response on this list. Second, VDBench is great for testing raw block i/o devices. I think a tool that does file system testing will get you better data. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] How to bind the oracle 9i data file to zfs volumes
Manoj Joseph wrote: Simon wrote: So,does mean this is oracle bug ? Or it's impossible(or inappropriate) to use ZFS/SVM volumes to create oracle data file,instead,should use zfs or ufs filesystem to do this. Oracle can use SVM volumes to hold its data. Unless I am mistaken, it should be able to use zvols as well. Yes. Though I believe most people will prefer regular file systems or ASM. We discuss performance implications on the ZFS wiki at http://solarisinternals.com However, googling for 'zvol + Oracle' did not get me anything useful. Perhaps it is not a configuration that is very popular. ;) Sounds like an opportunity... please share your experiences. -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
