Re: [zfs-discuss] Best practice for boot partition layout in ZFS
On 04/ 6/11 01:05 PM, Richard Elling wrote: On Apr 6, 2011, at 12:01 PM, Linder, Doug wrote: Torrey Minton wrote: I'm sure someone has a really good reason to keep /var separated but those cases are fewer and> far between than I saw 10 years ago. I agree that the causes and repercussions are less now than they were a long time ago. But /var still can and sometimes does fill up, and it is kind of handy to have quotas and separate filesystem settings and so on. I guess there's no overall crying reason to use a separate /var, but there's always this argument: it can't hurt anything. Especially with ZFS. In the old days if /var was a separate partition then you risked making it too big or too small. But given the flexibility of ZFS, I think the question is really "is there any reason *not* to put /var on a separate ZFS filesystem?" Yes. For backup/restore the unit of management is file system. More file systems results in more complicated backup/restore that increases RTO and costs. This was always the Achille's heel of separate /var. /var typically contains instance specific and instance shared data. Common examples of data that is shared between boot environments are mail and (if you have an install server), /var/ai. It may also be desirable for non-global zones to share portions of /var. I had posted a proposal to have a shared /var on caiman-discuss a while back. Many users responded and we are evaluating the feedback. -Sanjay -- richard ___ 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 to Solaris/FBSD/OpenIndiana/Nexenta VM guest?
On Wed, Apr 6, 2011 at 10:42 AM, Paul Kraus wrote: > I thought I saw that with zpool 10 (or was it 15) the zfs send > format had been committed and you *could* send/recv between different > version of zpool/zfs. From Solaris 10U9 with zpool 22 manpage for zfs: There is still a problem if the dataset version is too high. I *believe* that a 'zfs send -R' should send the zfs version, and that zfs receive will create any new datasets using that version. (I have a received dataset here that's zfs v 4, whereas everything else in the pool is v5.) As long as you don't do a zfs upgrade after that point, you should be fine. It's probably a good idea to check that the received versions are the same as the source before doing a destroy though. ;-) One other thing that I forgot to mention in my last mail too: If you're receiving into a VM, make sure that the VM can manage redundancy on its zfs storage, and not just multiple vdsk on the same host disk / lun. Either give it access to the raw devices, or use iSCSI, or create your vdsk on different luns and raidz them, etc. -B -- Brandon High : bh...@freaks.com ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Best practice for boot partition layout in ZFS
On Apr 6, 2011, at 12:01 PM, Linder, Doug wrote: > Torrey Minton wrote: > >> I'm sure someone has a really good reason to keep /var separated but those >> cases are fewer and > far between than I saw 10 years ago. > > I agree that the causes and repercussions are less now than they were a long > time ago. But /var still can and sometimes does fill up, and it is kind of > handy to have quotas and separate filesystem settings and so on. > > I guess there's no overall crying reason to use a separate /var, but there's > always this argument: it can't hurt anything. Especially with ZFS. In the > old days if /var was a separate partition then you risked making it too big > or too small. But given the flexibility of ZFS, I think the question is > really "is there any reason *not* to put /var on a separate ZFS filesystem?" Yes. For backup/restore the unit of management is file system. More file systems results in more complicated backup/restore that increases RTO and costs. This was always the Achille's heel of separate /var. -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Best practice for boot partition layout in ZFS
Torrey Minton wrote: > I'm sure someone has a really good reason to keep /var separated but those > cases are fewer and > far between than I saw 10 years ago. I agree that the causes and repercussions are less now than they were a long time ago. But /var still can and sometimes does fill up, and it is kind of handy to have quotas and separate filesystem settings and so on. I guess there's no overall crying reason to use a separate /var, but there's always this argument: it can't hurt anything. Especially with ZFS. In the old days if /var was a separate partition then you risked making it too big or too small. But given the flexibility of ZFS, I think the question is really "is there any reason *not* to put /var on a separate ZFS filesystem?" Doug Linder -- Learn more about Merchant Link at www.merchantlink.com. THIS MESSAGE IS CONFIDENTIAL. This e-mail message and any attachments are proprietary and confidential information intended only for the use of the recipient(s) named above. If you are not the intended recipient, you may not print, distribute, or copy this message or any attachments. If you have received this communication in error, please notify the sender by return e-mail and delete this message and any attachments from your computer. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS send/receive to Solaris/FBSD/OpenIndiana/Nexenta VM guest?
On 04/ 6/11 11:42 AM, Paul Kraus wrote: On Wed, Apr 6, 2011 at 1:14 PM, Brandon High wrote: The only thing to watch out for is to make sure that the receiving datasets aren't a higher version that the zfs version that you'll be using on the replacement server. Because you can't downgrade a dataset, using snv_151a and planning to send to Nexenta as a final step will trip you up unless you explicitly create them with a lower version. I thought I saw that with zpool 10 (or was it 15) the zfs send format had been committed and you *could* send/recv between different version of zpool/zfs. From Solaris 10U9 with zpool 22 manpage for zfs: The format of the stream is committed. You will be able to receive your streams on future versions of ZFS. correct. -or- does this just mean upward compatibility ? In other words I can send from pool 15 to pool 22 but not the other way around. It does mean upward compatibility only, but I believe that it's the dataset version that matters, not the pool version, and the dataset version has not changed as often as the pool version: root@v40z-brm-02:/home/lalt/ztest# zfs get version rpool/export/home NAME PROPERTY VALUESOURCE rpool/export/home version 5- root@v40z-brm-02:/home/lalt/ztest# zpool get version rpool NAME PROPERTY VALUESOURCE rpool version 32 default (someone still on the zfs team please correct me if that's wrong.) Lori ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] cannot destroy snapshot
On 04/06/11 12:43, Paul Kraus wrote: xxx> zfs holds zpool-01/dataset-01@1299636001 NAME TAGTIMESTAMP zpool-01/dataset-01@1299636001 .send-18440-0 Tue Mar 15 20:00:39 2011 xxx> zfs holds zpool-01/dataset-01@1300233615 NAME TAGTIMESTAMP zpool-01/dataset-01@1300233615 .send-18440-0 Tue Mar 15 20:00:47 2011 xxx> That is what I was looking for. Looks like when a zfs send got killed it left a hanging lock (hold) around. I assume the next export/import (not likely as this is a production zpool) or a reboot (will happen eventually, and I can wait) these will clear. Unless there is a way to force clear the hold. The user holds won't be released by an export/import or a reboot. "zfs get defer_destroy snapname" will show whether this snapshot is marked for deferred destroy and "zfs release .send-18440-0 snapname" will clear that hold. If the snapshot is marked for deferred destroy then the release of the last tag will also destroy it. -- Rich ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS send/receive to Solaris/FBSD/OpenIndiana/Nexenta VM guest?
On Wed, Apr 6, 2011 at 1:14 PM, Brandon High wrote: > The only thing to watch out for is to make sure that the receiving datasets > aren't a higher version that the zfs version that you'll be using on the > replacement server. Because you can't downgrade a dataset, using snv_151a > and planning to send to Nexenta as a final step will trip you up unless you > explicitly create them with a lower version. I thought I saw that with zpool 10 (or was it 15) the zfs send format had been committed and you *could* send/recv between different version of zpool/zfs. From Solaris 10U9 with zpool 22 manpage for zfs: The format of the stream is committed. You will be able to receive your streams on future versions of ZFS. -or- does this just mean upward compatibility ? In other words I can send from pool 15 to pool 22 but not the other way around. -- {1-2-3-4-5-6-7-} Paul Kraus -> Senior Systems Architect, Garnet River ( http://www.garnetriver.com/ ) -> Sound Coordinator, Schenectady Light Opera Company ( http://www.sloctheater.org/ ) -> Technical Advisor, RPI Players ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS send/receive to Solaris/FBSD/OpenIndiana/Nexenta VM guest?
On Tue, Apr 5, 2011 at 12:38 PM, Joe Auty wrote: > How about getting a little more crazy... What if this entire server > temporarily hosting this data was a VM guest running ZFS? I don't foresee > this being a problem either, but with so > The only thing to watch out for is to make sure that the receiving datasets aren't a higher version that the zfs version that you'll be using on the replacement server. Because you can't downgrade a dataset, using snv_151a and planning to send to Nexenta as a final step will trip you up unless you explicitly create them with a lower version. -B -- Brandon High : bh...@freaks.com ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS send/receive to Solaris/FBSD/OpenIndiana/Nexenta VM guest?
On Wed, Apr 6, 2011 at 10:51 AM, David Dyer-Bennet wrote: > > On Tue, April 5, 2011 14:38, Joe Auty wrote: >> Also, more generally, is ZFS send/receive mature enough that when you do >> data migrations you don't stress about this? Piece of cake? The >> difficulty of this whole undertaking will influence my decision and the >> whole timing of all of this. > > A full send / receive has been reliable for a long time. With a real > (large) data set, it's often a long run. It's often done over a network, > and any network outage can break the run, and at that point you start > over, which can be annoying. If the servers themselves can't stay up for > 10 or 20 hours you presumably aren't ready to put them into production > anyway :-). At my employer we have about 20TB of data in one city and a zfs replicated copy of it in another city. The data is spread out over 15 pools and over 200 datasets. The initial full replication of the larger datasets (the largest is 3 TB) took days, the largest even took close to two weeks. The incremental send/recv sessions are much quicker, based on how much data has changed, but we run the replication script every 4 hours and it usually completes before the next scheduled run. Once we got past a few bugs in both my script and the older zfs code (we are at zpool 22 and zfs 4 right now, we started all this at zpool 10) the replications have been flawless. >> I'm also thinking that a ZFS VM guest might be a nice way to maintain a >> remote backup of this data, if I can install the VM image on a >> drive/partition large enough to house my data. This seems like it would >> be a little less taxing than rsync cronjobs? > > I'm a big fan of rsync, in cronjobs or wherever. What it won't do is > properly preserve ZFS ACLs, and ZFS snapshots, though. I moved from using > rsync to using zfs send/receive for my backup scheme at home, and had > considerable trouble getting that all working (using incremental > send/receive when there are dozens of snapshots new since last time). But > I did eventually get up to recent enough code that it's working reliably > now. We went with zfs send/recv over rsync for two big reasons, an incremental zfs send is much, much faster than an rsync if you have lots of files (our 20TB of data consists of 200 million files), and we are leveraging zfs ACLs and need them preserved on the copy. I have not tried zfs on a VM guest. -- {1-2-3-4-5-6-7-} Paul Kraus -> Senior Systems Architect, Garnet River ( http://www.garnetriver.com/ ) -> Sound Coordinator, Schenectady Light Opera Company ( http://www.sloctheater.org/ ) -> Technical Advisor, RPI Players ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] cannot destroy snapshot
On Tue, Apr 5, 2011 at 9:26 PM, Edward Ned Harvey wrote: > This may not apply to you, but in some other unrelated situation it was > useful... > > Try zdb -d poolname > In an older version of zpool, under certain conditions, there would > sometimes be "hidden" clones listed with a % in the name. Maybe the % won't > be there in your case, but maybe you have some other manifestation of the > hidden clone problem? I have seen the dataset with a '%' in the name, but that was during a zfs recv (and if the zfs recv dies, then it sometimes hangs around and has to be destroyed, and the zfs destroy claims to fail even though it succeeds ;-), but not in this case. The snapshots are all valid (I just can't destroy two of them), we are snapshotting on a fairly frequent basis as we are loading data. Thanks for the suggestion. xxx> zdb -d zpool-01 Dataset mos [META], ID 0, cr_txg 4, 18.7G, 745 objects Dataset zpool-01/dataset-01@1302019202 [ZPL], ID 140, cr_txg 654658, 38.9G, 990842 objects Dataset zpool-01/dataset-01@1302051600 [ZPL], ID 158, cr_txg 655776, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1302062401 [ZPL], ID 189, cr_txg 656162, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1301951162 [ZPL], ID 108, cr_txg 652292, 1.02M, 478 objects Dataset zpool-01/dataset-01@1302087601 [ZPL], ID 254, cr_txg 657065, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1302105601 [ZPL], ID 291, cr_txg 657710, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1302058800 [ZPL], ID 164, cr_txg 656033, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1299636001 [ZPL], ID 48, cr_txg 560375, 1.12T, 28468324 objects Dataset zpool-01/dataset-01@1302007173 [ZPL], ID 125, cr_txg 654202, 1.09M, 506 objects Dataset zpool-01/dataset-01@1302055201 [ZPL], ID 161, cr_txg 655905, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1302080401 [ZPL], ID 248, cr_txg 656807, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1302044400 [ZPL], ID 152, cr_txg 655518, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1301950939 [ZPL], ID 106, cr_txg 652280, 1.02M, 478 objects Dataset zpool-01/dataset-01@1302015602 [ZPL], ID 137, cr_txg 654530, 10.3G, 175879 objects Dataset zpool-01/dataset-01@1302030001 [ZPL], ID 143, cr_txg 655029, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1300233615 [ZPL], ID 79, cr_txg 594951, 4.48T, 99259515 objects Dataset zpool-01/dataset-01@1302094801 [ZPL], ID 282, cr_txg 657323, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1302066001 [ZPL], ID 214, cr_txg 656291, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1302006933 [ZPL], ID 120, cr_txg 654181, 1.09M, 506 objects Dataset zpool-01/dataset-01@1302098401 [ZPL], ID 285, cr_txg 657452, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1302007755 [ZPL], ID 131, cr_txg 654240, 1.09M, 506 objects Dataset zpool-01/dataset-01@1302048001 [ZPL], ID 155, cr_txg 655647, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1302005414 [ZPL], ID 116, cr_txg 654119, 1.09M, 506 objects Dataset zpool-01/dataset-01@1302007469 [ZPL], ID 128, cr_txg 654221, 1.09M, 506 objects Dataset zpool-01/dataset-01@1302084001 [ZPL], ID 251, cr_txg 656936, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1302076801 [ZPL], ID 245, cr_txg 656678, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1302069601 [ZPL], ID 217, cr_txg 656420, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1302073201 [ZPL], ID 242, cr_txg 656549, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1302102001 [ZPL], ID 288, cr_txg 657581, 71.1G, 1845553 objects Dataset zpool-01/dataset-01@1302005162 [ZPL], ID 112, cr_txg 654101, 1.09M, 506 objects Dataset zpool-01/dataset-01@1302012001 [ZPL], ID 134, cr_txg 654391, 1.18G, 63312 objects Dataset zpool-01/dataset-01@1302004805 [ZPL], ID 110, cr_txg 654085, 1.09M, 506 objects Dataset zpool-01/dataset-01@1302006769 [ZPL], ID 118, cr_txg 654171, 1.09M, 506 objects Dataset zpool-01/dataset-01@1302091201 [ZPL], ID 257, cr_txg 657194, 71.1G, 1845553 objects Dataset zpool-01/dataset-01 [ZPL], ID 84, cr_txg 439406, 71.1G, 1845553 objects Dataset zpool-01 [ZPL], ID 16, cr_txg 1, 39.3K, 5 objects xxx> -- {1-2-3-4-5-6-7-} Paul Kraus -> Senior Systems Architect, Garnet River ( http://www.garnetriver.com/ ) -> Sound Coordinator, Schenectady Light Opera Company ( http://www.sloctheater.org/ ) -> Technical Advisor, RPI Players ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] cannot destroy snapshot
On Tue, Apr 5, 2011 at 6:56 PM, Rich Morris wrote: > On 04/05/11 17:29, Ian Collins wrote: > If there are clones then zfs destroy should report that. The error being > reported is "dataset is busy" which would be reported if there are user > holds on the snapshots that can't be deleted. > > Try running "zfs holds zpool-01/dataset-01@1299636001" xxx> zfs holds zpool-01/dataset-01@1299636001 NAME TAGTIMESTAMP zpool-01/dataset-01@1299636001 .send-18440-0 Tue Mar 15 20:00:39 2011 xxx> zfs holds zpool-01/dataset-01@1300233615 NAME TAGTIMESTAMP zpool-01/dataset-01@1300233615 .send-18440-0 Tue Mar 15 20:00:47 2011 xxx> That is what I was looking for. Looks like when a zfs send got killed it left a hanging lock (hold) around. I assume the next export/import (not likely as this is a production zpool) or a reboot (will happen eventually, and I can wait) these will clear. Unless there is a way to force clear the hold. Thanks Rich. -- {1-2-3-4-5-6-7-} Paul Kraus -> Senior Systems Architect, Garnet River ( http://www.garnetriver.com/ ) -> Sound Coordinator, Schenectady Light Opera Company ( http://www.sloctheater.org/ ) -> Technical Advisor, RPI Players ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Best practice for boot partition layout in ZFS
On Wed, April 6, 2011 11:29, Gary Mills wrote: > People forget (c), the ability to set different filesystem options on > /var. You might want to have `setuid=off' for improved security, for > example. Or better yet: exec=off,devices=off. Another handy one could be "compression=on" (or a even "gzip-[1-9]"). ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS send/receive to Solaris/FBSD/OpenIndiana/Nexenta VM guest?
On Wed, April 6, 2011 10:51, David Dyer-Bennet wrote: > I'm a big fan of rsync, in cronjobs or wherever. What it won't do is > properly preserve ZFS ACLs, and ZFS snapshots, though. I moved from using > rsync to using zfs send/receive for my backup scheme at home, and had > considerable trouble getting that all working (using incremental > send/receive when there are dozens of snapshots new since last time). But > I did eventually get up to recent enough code that it's working reliably > now. You may be interested in these scripts: http://www.freshports.org/sysutils/zfs-replicate/ http://www.freshports.org/sysutils/zxfer/ Not sure how FreeBSD-specific these are, but one was originally written for (Open)Solaris AFAICT. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Best practice for boot partition layout in ZFS
On 4/6/2011 11:08 AM, Erik Trimble wrote: Traditionally, the reason for a separate /var was one of two major items: (a) /var was writable, and / wasn't - this was typical of diskless or minimal local-disk configurations. Modern packaging systems are making this kind of configuration increasingly difficult. (b) /var held a substantial amount of data, which needed to be handled separately from / - mail and news servers are a classic example For typical machines nowdays, with large root disks, there is very little chance of /var suddenly exploding and filling / (the classic example of being screwed... ). Outside of the above two cases, about the only other place I can see that having /var separate is a good idea is for certain test machines, where you expect frequent memory dumps (in /var/crash) - if you have a large amount of RAM, you'll need a lot of disk space, so it might be good to limit /var in this case by making it a separate dataset. Some more info ala (b) - The "something filled up the root fs and the box crashed" problem was fixed for awhile ago. It's still a drag cleaning up an errant process that is filling up a file system but it shouldn't crash/panic anymore. However, old habits die hard, especially at government sites where the rules require a papal bull to be changed, so I think the option was left to keep folks happy more than any practical reason. I'm sure someone has a really good reason to keep /var separated but those cases are fewer and far between than I saw 10 years ago. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Best practice for boot partition layout in ZFS
On Wed, Apr 06, 2011 at 08:08:06AM -0700, Erik Trimble wrote: >On 4/6/2011 7:50 AM, Lori Alt wrote: >On 04/ 6/11 07:59 AM, Arjun YK wrote: > >I'm not sure there's a defined "best practice". Maybe someone else >can answer that question. My guess is that in environments where, >before, a separate ufs /var slice was used, a separate zfs /var >dataset with a quota might now be appropriate. >Lori > >Traditionally, the reason for a separate /var was one of two major >items: >(a) /var was writable, and / wasn't - this was typical of diskless or >minimal local-disk configurations. Modern packaging systems are making >this kind of configuration increasingly difficult. >(b) /var held a substantial amount of data, which needed to be handled >separately from / - mail and news servers are a classic example >For typical machines nowdays, with large root disks, there is very >little chance of /var suddenly exploding and filling / (the classic >example of being screwed... ). Outside of the above two cases, >about the only other place I can see that having /var separate is a >good idea is for certain test machines, where you expect frequent >memory dumps (in /var/crash) - if you have a large amount of RAM, >you'll need a lot of disk space, so it might be good to limit /var in >this case by making it a separate dataset. People forget (c), the ability to set different filesystem options on /var. You might want to have `setuid=off' for improved security, for example. -- -Gary Mills--Unix Group--Computer and Network Services- ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Best practice for boot partition layout in ZFS
On 4/6/2011 7:50 AM, Lori Alt wrote: On 04/ 6/11 07:59 AM, Arjun YK wrote: Hi, I am trying to use ZFS for boot, and kind of confused about how the boot paritions like /var to be layed out. With old UFS, we create /var as sepearate filesystem to avoid various logs filling up the / filesystem I believe that creating /var as a separate file system was a common practice, but not a universal one. It really depended on the environment and local requirements. With ZFS, during the OS install it gives the option to "Put /var on a separate dataset", but no option is given to set quota. May be, others set quota manually. Having a separate /var dataset gives you the option of setting a quota on it later. That's why we provided the option. It was a way of enabling administrators to get the same effect as having a separate /var slice did with ufs. Administrators can choose to use it or not, depending on local requirements. So, I am trying to understand what's the best practice for /var in ZFS. Is that exactly same as in UFS or is there anything different ? I'm not sure there's a defined "best practice". Maybe someone else can answer that question. My guess is that in environments where, before, a separate ufs /var slice was used, a separate zfs /var dataset with a quota might now be appropriate. Lori Could someone share some thoughts ? Thanks Arjun Traditionally, the reason for a separate /var was one of two major items: (a) /var was writable, and / wasn't - this was typical of diskless or minimal local-disk configurations. Modern packaging systems are making this kind of configuration increasingly difficult. (b) /var held a substantial amount of data, which needed to be handled separately from / - mail and news servers are a classic example For typical machines nowdays, with large root disks, there is very little chance of /var suddenly exploding and filling / (the classic example of being screwed... ). Outside of the above two cases, about the only other place I can see that having /var separate is a good idea is for certain test machines, where you expect frequent memory dumps (in /var/crash) - if you have a large amount of RAM, you'll need a lot of disk space, so it might be good to limit /var in this case by making it a separate dataset. -- 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 send/receive to Solaris/FBSD/OpenIndiana/Nexenta VM guest?
On Tue, April 5, 2011 14:38, Joe Auty wrote: > Migrating to a new machine I understand is a simple matter of ZFS > send/receive, but reformatting the existing drives to host my existing > data is an area I'd like to learn a little more about. In the past I've > asked about this and was told that it is possible to do a send/receive > to accommodate this, and IIRC this doesn't have to be to a ZFS server > with the same number of physical drives? The internal structure of the pool (how many vdevs, and what kind) is irrelevant to zfs send / receive. So I routinely send from a pool of 3 mirrored pairs of disks to a pool of one large drive, for example (it's how I do my backups). I've also gone the other way once :-( (It's good to have backups). I'm not 100.00% sure I understand what you're asking; does that answer it? Mind you, this can be slow. On my little server (under 1TB filled) the full backup takes about 7 hours (largely because the single large external drive is a USB drive; the bottleneck is the USB). Luckily an incremental backup is rather faster. > How about getting a little more crazy... What if this entire server > temporarily hosting this data was a VM guest running ZFS? I don't > foresee this being a problem either, but with so much at stake I thought > I would double check :) When I say temporary I mean simply using this > machine as a place to store the data long enough to wipe the original > server, install the new OS to the original server, and restore the data > using this VM as the data source. I haven't run ZFS extensively in VMs (mostly just short-lived small test setups). From my limited experience, and what I've heard on the list, it's solid and reliable, though, which is what you need for that application. > Also, more generally, is ZFS send/receive mature enough that when you do > data migrations you don't stress about this? Piece of cake? The > difficulty of this whole undertaking will influence my decision and the > whole timing of all of this. A full send / receive has been reliable for a long time. With a real (large) data set, it's often a long run. It's often done over a network, and any network outage can break the run, and at that point you start over, which can be annoying. If the servers themselves can't stay up for 10 or 20 hours you presumably aren't ready to put them into production anyway :-). > I'm also thinking that a ZFS VM guest might be a nice way to maintain a > remote backup of this data, if I can install the VM image on a > drive/partition large enough to house my data. This seems like it would > be a little less taxing than rsync cronjobs? I'm a big fan of rsync, in cronjobs or wherever. What it won't do is properly preserve ZFS ACLs, and ZFS snapshots, though. I moved from using rsync to using zfs send/receive for my backup scheme at home, and had considerable trouble getting that all working (using incremental send/receive when there are dozens of snapshots new since last time). But I did eventually get up to recent enough code that it's working reliably now. If you can provision big enough data stores for your VM to hold what you need, that seems a reasonable approach to me, but I haven't tried anything much like it, so my opinion is, if you're very lucky, maybe worth what you paid for it. -- David Dyer-Bennet, d...@dd-b.net; http://dd-b.net/ Snapshots: http://dd-b.net/dd-b/SnapshotAlbum/data/ Photos: http://dd-b.net/photography/gallery/ Dragaera: http://dragaera.info ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Best practice for boot partition layout in ZFS
On 04/ 6/11 07:59 AM, Arjun YK wrote: Hi, I am trying to use ZFS for boot, and kind of confused about how the boot paritions like /var to be layed out. With old UFS, we create /var as sepearate filesystem to avoid various logs filling up the / filesystem I believe that creating /var as a separate file system was a common practice, but not a universal one. It really depended on the environment and local requirements. With ZFS, during the OS install it gives the option to "Put /var on a separate dataset", but no option is given to set quota. May be, others set quota manually. Having a separate /var dataset gives you the option of setting a quota on it later. That's why we provided the option. It was a way of enabling administrators to get the same effect as having a separate /var slice did with ufs. Administrators can choose to use it or not, depending on local requirements. So, I am trying to understand what's the best practice for /var in ZFS. Is that exactly same as in UFS or is there anything different ? I'm not sure there's a defined "best practice". Maybe someone else can answer that question. My guess is that in environments where, before, a separate ufs /var slice was used, a separate zfs /var dataset with a quota might now be appropriate. Lori Could someone share some thoughts ? Thanks Arjun ___ 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
[zfs-discuss] ZFS send/receive to Solaris/FBSD/OpenIndiana/Nexenta VM guest?
Hello, I'm debating an OS change and also thinking about my options for data migration to my next server, whether it is on new or the same hardware. Migrating to a new machine I understand is a simple matter of ZFS send/receive, but reformatting the existing drives to host my existing data is an area I'd like to learn a little more about. In the past I've asked about this and was told that it is possible to do a send/receive to accommodate this, and IIRC this doesn't have to be to a ZFS server with the same number of physical drives? How about getting a little more crazy... What if this entire server temporarily hosting this data was a VM guest running ZFS? I don't foresee this being a problem either, but with so much at stake I thought I would double check :) When I say temporary I mean simply using this machine as a place to store the data long enough to wipe the original server, install the new OS to the original server, and restore the data using this VM as the data source. Also, more generally, is ZFS send/receive mature enough that when you do data migrations you don't stress about this? Piece of cake? The difficulty of this whole undertaking will influence my decision and the whole timing of all of this. I'm also thinking that a ZFS VM guest might be a nice way to maintain a remote backup of this data, if I can install the VM image on a drive/partition large enough to house my data. This seems like it would be a little less taxing than rsync cronjobs? -- Joe Auty, NetMusician NetMusician helps musicians, bands and artists create beautiful, professional, custom designed, career-essential websites that are easy to maintain and to integrate with popular social networks. www.netmusician.org j...@netmusician.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Best practice for boot partition layout in ZFS
Hi, I am trying to use ZFS for boot, and kind of confused about how the boot paritions like /var to be layed out. With old UFS, we create /var as sepearate filesystem to avoid various logs filling up the / filesystem With ZFS, during the OS install it gives the option to "Put /var on a separate dataset", but no option is given to set quota. May be, others set quota manually. So, I am trying to understand what's the best practice for /var in ZFS. Is that exactly same as in UFS or is there anything different ? Could someone share some thoughts ? Thanks Arjun ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss