Roman Ivanov
2009-Dec-12 18:05 UTC
[zfs-discuss] all zfs snapshot made by TimeSlider destroyed after upgrading to b129
Am I missing something? I have had monthly,weekly,daily,hourly,frequent snapshots since March 2009. Now with new b129 I lost all of them.>From zpool history:2009-12-12.20:30:02 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:weekly-2009-11-26-09:28 2009-12-12.20:30:03 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:weekly-2009-11-18-23:37 2009-12-12.20:30:04 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:monthly-2009-10-17-20:32 2009-12-12.20:30:04 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-12-19:47 2009-12-12.20:30:05 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-15:59 2009-12-12.20:30:05 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-14:54 2009-12-12.20:30:06 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-13:54 2009-12-12.20:30:07 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-12:54 2009-12-12.20:30:07 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-11:54 ..... 2009-12-12.20:30:43 zfs destroy -r rpool/rixxxxxx at zfs-auto-snap:monthly-2009-06-16-08:15 2009-12-12.20:30:44 zfs destroy -r rpool/rixxxxxx at zfs-auto-snap:monthly-2009-05-16-11:52 2009-12-12.20:30:44 zfs destroy -r rpool/rixxxxxx at zfs-auto-snap:monthly-2009-04-16-08:06 2009-12-12.20:30:46 zfs destroy -r rpool/rixxxxxx at zfs-auto-snap:monthly-2009-03-16-18:55 Current zfs list -t all: NAME USED AVAIL REFER MOUNTPOINT rpool 54,3G 83,5G 63,5K /rpool rpool/ROOT 17,1G 83,5G 18K legacy rpool/ROOT/b128a 28,5M 83,5G 9,99G legacy rpool/ROOT/b128a at zfs-auto-snap:frequent-2009-12-12-20:17 9,70M - 9,99G - rpool/ROOT/b129 17,1G 83,5G 10,2G legacy rpool/ROOT/b129 at 2009-09-04-11:28:13 3,74G - 10,0G - rpool/ROOT/b129 at zfs-auto-snap:weekly-2009-12-03-14:59 1,25G - 10,2G - rpool/ROOT/b129 at zfs-auto-snap:weekly-2009-12-10-14:59 550M - 10,4G - rpool/ROOT/b129 at 2009-12-12-17:11:35 29,9M - 10,4G - rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-12-21:00 0 - 10,2G - rpool/ROOT/b129 at zfs-auto-snap:-2009-12-12-21:00 0 - 10,2G - rpool/dump 1023M 83,5G 1023M - rpool/rixxxxxx 35,2G 83,5G 34,9G /export/home/rixxxxxx rpool/rixxxxxx at zfs-auto-snap:weekly-2009-12-03-14:59 190M - 31,8G - rpool/rixxxxxx at zfs-auto-snap:weekly-2009-12-10-14:59 116M - 34,9G - rpool/rixxxxxx at zfs-auto-snap:-2009-12-12-21:00 2,29M - 34,9G - rpool/swap 1023M 84,3G 275M - The latest snapshot does not have word "frequent" in it. Moreover hourly snapshot died right after born 2009-12-12.21:00:02 zfs snapshot -r rpool/rixxxxxx at zfs-auto-snap:hourly-2009-12-12-21:00 2009-12-12.21:00:02 zfs destroy -r rpool/rixxxxxx at zfs-auto-snap:hourly-2009-12-12-21:00 2009-12-12.21:00:03 zfs destroy -r rpool/rixxxxxx at zfs-auto-snap:-2009-12-12-20:45 2009-12-12.21:00:03 zfs snapshot -r rpool/rixxxxxx at zfs-auto-snap:-2009-12-12-21:00 -- This message posted from opensolaris.org
Daniel Carosone
2009-Dec-14 03:31 UTC
[zfs-discuss] all zfs snapshot made by TimeSlider destroyed after upgrading to b129
I can''t (yet!) say I''ve seen the same, with respect to disappearing snapshots. However, I can confirm that I am seeing the same thing, with respect to snapshots without the "frequent" prefix.. $ zfs list -t snapshot | fgrep :- rpool at zfs-auto-snap:-2009-12-14-13:15 0 - 83.5K - rpool/ROOT at zfs-auto-snap:-2009-12-14-13:15 0 - 19K - rpool/ROOT/opensolaris at zfs-auto-snap:-2009-12-14-13:15 0 - 3.78G - rpool/ROOT/opensolaris-1 at zfs-auto-snap:-2009-12-14-13:15 0 - 4.81G - rpool/ROOT/opensolaris-2 at zfs-auto-snap:-2009-12-14-13:15 0 - 5.65G - rpool/ROOT/snv_129 at zfs-auto-snap:-2009-12-14-13:15 478K - 5.75G - rpool/export at zfs-auto-snap:-2009-12-14-13:15 0 - 24K - rpool/export/VirtualBox at zfs-auto-snap:-2009-12-14-13:15 0 - 18.3G - rpool/export/home at zfs-auto-snap:-2009-12-14-13:15 0 - 23K - rpool/export/home/dan at zfs-auto-snap:-2009-12-14-13:15 0 - 1.80G - rpool/var_postgres at zfs-auto-snap:-2009-12-14-13:15 56K - 39.2M - -- This message posted from opensolaris.org
Ross
2009-Dec-14 10:23 UTC
[zfs-discuss] all zfs snapshot made by TimeSlider destroyed after upgrading to b129
There was an announcement made in November about auto snapshots being made obsolete in build 128, I assume major changes are afoot: http://www.opensolaris.org/jive/thread.jspa?messageID=437516&tstart=0#437516 -- This message posted from opensolaris.org
Daniel Carosone
2009-Dec-14 12:20 UTC
[zfs-discuss] all zfs snapshot made by TimeSlider destroyed after upgrading to b129
> There was an announcement made in November about auto > snapshots being made obsolete in build 128That thread (which I know well) talks about the replacement of the [b]implementation[/b], while retaining the (majority of) the behaviour and configuration interface. The old implementation had plenty of issues. It seems inevitably the new implementation comes with a few new bugs, too. -- This message posted from opensolaris.org
Cindy Swearingen
2009-Dec-15 15:36 UTC
[zfs-discuss] all zfs snapshot made by TimeSlider destroyed after upgrading to b129
Hi-- I haven''t had a chance to reproduce this problem but Niall''s heads up message, says that default schedules that include "frequent" still work: http://mail.opensolaris.org/pipermail/zfs-auto-snapshot/2009-November/000199.html I included a snippet of his instructions below. If this doesn''t help, I''ll see if Niall can comment. Thanks, Cindy ***************************************** For those who want to use time-slider without going through the GUI, simply enable/configure (or create) the auto-snapshot instances you need then enable the time-slider SMF service. time-slider will pick up the enabled auto-snapshot instances and schedule snapshots for them. For folks who prefer to continue using zfs-auto-snapshot, you will need to remove SUNWgnome-time-slider and install the existing zfs-auto-snapshot packages instead. On 12/12/09 11:05, Roman Ivanov wrote:> Am I missing something? > > I have had monthly,weekly,daily,hourly,frequent snapshots since March 2009. > Now with new b129 I lost all of them. >>From zpool history: > > 2009-12-12.20:30:02 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:weekly-2009-11-26-09:28 > 2009-12-12.20:30:03 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:weekly-2009-11-18-23:37 > 2009-12-12.20:30:04 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:monthly-2009-10-17-20:32 > 2009-12-12.20:30:04 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-12-19:47 > 2009-12-12.20:30:05 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-15:59 > 2009-12-12.20:30:05 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-14:54 > 2009-12-12.20:30:06 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-13:54 > 2009-12-12.20:30:07 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-12:54 > 2009-12-12.20:30:07 zfs destroy -r rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-11:54 > ..... > 2009-12-12.20:30:43 zfs destroy -r rpool/rixxxxxx at zfs-auto-snap:monthly-2009-06-16-08:15 > 2009-12-12.20:30:44 zfs destroy -r rpool/rixxxxxx at zfs-auto-snap:monthly-2009-05-16-11:52 > 2009-12-12.20:30:44 zfs destroy -r rpool/rixxxxxx at zfs-auto-snap:monthly-2009-04-16-08:06 > 2009-12-12.20:30:46 zfs destroy -r rpool/rixxxxxx at zfs-auto-snap:monthly-2009-03-16-18:55 > > Current zfs list -t all: > NAME USED AVAIL REFER MOUNTPOINT > rpool 54,3G 83,5G 63,5K /rpool > rpool/ROOT 17,1G 83,5G 18K legacy > rpool/ROOT/b128a 28,5M 83,5G 9,99G legacy > rpool/ROOT/b128a at zfs-auto-snap:frequent-2009-12-12-20:17 9,70M - 9,99G - > rpool/ROOT/b129 17,1G 83,5G 10,2G legacy > rpool/ROOT/b129 at 2009-09-04-11:28:13 3,74G - 10,0G - > rpool/ROOT/b129 at zfs-auto-snap:weekly-2009-12-03-14:59 1,25G - 10,2G - > rpool/ROOT/b129 at zfs-auto-snap:weekly-2009-12-10-14:59 550M - 10,4G - > rpool/ROOT/b129 at 2009-12-12-17:11:35 29,9M - 10,4G - > rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-12-21:00 0 - 10,2G - > rpool/ROOT/b129 at zfs-auto-snap:-2009-12-12-21:00 0 - 10,2G - > rpool/dump 1023M 83,5G 1023M - > rpool/rixxxxxx 35,2G 83,5G 34,9G /export/home/rixxxxxx > rpool/rixxxxxx at zfs-auto-snap:weekly-2009-12-03-14:59 190M - 31,8G - > rpool/rixxxxxx at zfs-auto-snap:weekly-2009-12-10-14:59 116M - 34,9G - > rpool/rixxxxxx at zfs-auto-snap:-2009-12-12-21:00 2,29M - 34,9G - > rpool/swap 1023M 84,3G 275M - > > The latest snapshot does not have word "frequent" in it. Moreover hourly snapshot died right after born > 2009-12-12.21:00:02 zfs snapshot -r rpool/rixxxxxx at zfs-auto-snap:hourly-2009-12-12-21:00 > 2009-12-12.21:00:02 zfs destroy -r rpool/rixxxxxx at zfs-auto-snap:hourly-2009-12-12-21:00 > 2009-12-12.21:00:03 zfs destroy -r rpool/rixxxxxx at zfs-auto-snap:-2009-12-12-20:45 > 2009-12-12.21:00:03 zfs snapshot -r rpool/rixxxxxx at zfs-auto-snap:-2009-12-12-21:00
Cindy Swearingen
2009-Dec-15 22:48 UTC
[zfs-discuss] all zfs snapshot made by TimeSlider destroyed after upgrading to b129
Okay, not much help. I found a couple more problems reported with workarounds, here: http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6904417 time-slider unable to start after upgrade to snv_128 http://defect.opensolaris.org/bz/show_bug.cgi?id=13301 time-slider ignores all filesystems in some pools The workaround in 13301 says this: Having upgraded to build 129, I now find that time-slider seems not to work at all. This gets it working again: # svccfg import /var/svc/manifest/system/filesystem/auto-snapshot-roleadd.xml Cindy On 12/15/09 08:36, Cindy Swearingen wrote:> Hi-- > > I haven''t had a chance to reproduce this problem but Niall''s heads up > message, says that default schedules that include "frequent" still > work: > > http://mail.opensolaris.org/pipermail/zfs-auto-snapshot/2009-November/000199.html > > > I included a snippet of his instructions below. > > If this doesn''t help, I''ll see if Niall can comment. > > Thanks, > > Cindy > > ***************************************** > > For those who want to use time-slider without going through the GUI, simply > enable/configure (or create) the auto-snapshot instances you need then > enable > the time-slider SMF service. time-slider will pick up the enabled > auto-snapshot > instances and schedule snapshots for them. > > For folks who prefer to continue using zfs-auto-snapshot, you will need to > remove SUNWgnome-time-slider and install the existing zfs-auto-snapshot > packages instead. > > > > On 12/12/09 11:05, Roman Ivanov wrote: >> Am I missing something? >> >> I have had monthly,weekly,daily,hourly,frequent snapshots since March >> 2009. >> Now with new b129 I lost all of them. >>> From zpool history: >> >> 2009-12-12.20:30:02 zfs destroy -r >> rpool/ROOT/b129 at zfs-auto-snap:weekly-2009-11-26-09:28 >> 2009-12-12.20:30:03 zfs destroy -r >> rpool/ROOT/b129 at zfs-auto-snap:weekly-2009-11-18-23:37 >> 2009-12-12.20:30:04 zfs destroy -r >> rpool/ROOT/b129 at zfs-auto-snap:monthly-2009-10-17-20:32 >> 2009-12-12.20:30:04 zfs destroy -r >> rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-12-19:47 >> 2009-12-12.20:30:05 zfs destroy -r >> rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-15:59 >> 2009-12-12.20:30:05 zfs destroy -r >> rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-14:54 >> 2009-12-12.20:30:06 zfs destroy -r >> rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-13:54 >> 2009-12-12.20:30:07 zfs destroy -r >> rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-12:54 >> 2009-12-12.20:30:07 zfs destroy -r >> rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-11-11:54 >> ..... >> 2009-12-12.20:30:43 zfs destroy -r >> rpool/rixxxxxx at zfs-auto-snap:monthly-2009-06-16-08:15 >> 2009-12-12.20:30:44 zfs destroy -r >> rpool/rixxxxxx at zfs-auto-snap:monthly-2009-05-16-11:52 >> 2009-12-12.20:30:44 zfs destroy -r >> rpool/rixxxxxx at zfs-auto-snap:monthly-2009-04-16-08:06 >> 2009-12-12.20:30:46 zfs destroy -r >> rpool/rixxxxxx at zfs-auto-snap:monthly-2009-03-16-18:55 >> >> Current zfs list -t all: >> NAME USED >> AVAIL REFER MOUNTPOINT >> rpool 54,3G >> 83,5G 63,5K /rpool >> rpool/ROOT 17,1G >> 83,5G 18K legacy >> rpool/ROOT/b128a 28,5M >> 83,5G 9,99G legacy >> rpool/ROOT/b128a at zfs-auto-snap:frequent-2009-12-12-20:17 9,70M >> - 9,99G - >> rpool/ROOT/b129 17,1G >> 83,5G 10,2G legacy >> rpool/ROOT/b129 at 2009-09-04-11:28:13 3,74G >> - 10,0G - >> rpool/ROOT/b129 at zfs-auto-snap:weekly-2009-12-03-14:59 1,25G >> - 10,2G - >> rpool/ROOT/b129 at zfs-auto-snap:weekly-2009-12-10-14:59 550M >> - 10,4G - >> rpool/ROOT/b129 at 2009-12-12-17:11:35 29,9M >> - 10,4G - >> rpool/ROOT/b129 at zfs-auto-snap:hourly-2009-12-12-21:00 0 >> - 10,2G - >> rpool/ROOT/b129 at zfs-auto-snap:-2009-12-12-21:00 0 >> - 10,2G - >> rpool/dump 1023M >> 83,5G 1023M - >> rpool/rixxxxxx 35,2G >> 83,5G 34,9G /export/home/rixxxxxx >> rpool/rixxxxxx at zfs-auto-snap:weekly-2009-12-03-14:59 190M >> - 31,8G - >> rpool/rixxxxxx at zfs-auto-snap:weekly-2009-12-10-14:59 116M >> - 34,9G - >> rpool/rixxxxxx at zfs-auto-snap:-2009-12-12-21:00 2,29M >> - 34,9G - >> rpool/swap 1023M >> 84,3G 275M - >> >> The latest snapshot does not have word "frequent" in it. Moreover >> hourly snapshot died right after born >> 2009-12-12.21:00:02 zfs snapshot -r >> rpool/rixxxxxx at zfs-auto-snap:hourly-2009-12-12-21:00 >> 2009-12-12.21:00:02 zfs destroy -r >> rpool/rixxxxxx at zfs-auto-snap:hourly-2009-12-12-21:00 >> 2009-12-12.21:00:03 zfs destroy -r >> rpool/rixxxxxx at zfs-auto-snap:-2009-12-12-20:45 >> 2009-12-12.21:00:03 zfs snapshot -r >> rpool/rixxxxxx at zfs-auto-snap:-2009-12-12-21:00 > _______________________________________________ > zfs-discuss mailing list > zfs-discuss at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Daniel Carosone
2009-Dec-16 00:09 UTC
[zfs-discuss] all zfs snapshot made by TimeSlider destroyed after upgrading to b129
None of these look like the issue either. With 128, I did have to edit the code to avoid the month rollover error, and add the missing dependency dbus-python26. I think I have a new install that went to 129 without having auto snapshots enabled yet. When I can get to that machine later, I will enable them there and see whether the same fault is apparent, in case it''s some kind of compatibility problem with older state. Also not much help, sorry.. I don''t have an opportunity to spend time digging into it much further just now. -- This message posted from opensolaris.org