On Tue, Jan 22, 2008 at 10:14:34AM -0800, Dan Anderson wrote: > I had a problem with the NV79a smb/server failing, and here's what I > did to restart it. Having a root cause would be nice, but at least > this gets the service up and running again.
Sure. > # svcs -l idmap > ... > state maintenance > ... > > tail /var/svc/log/system-idmap:default.log > Database error during BEGIN TRANSACTION; (database is full) > [ Jan 21 10:36:50 Stopping because process dumped core. ] Is this *all* that was in the log? Which DB is full? If /var/idmap/idmap.db then is /var full? If /var/run/idmap/idmap.db then is the system short of virtual memory? On Tue, 22 Jan 2008 13:16:45 -0800, Dan Anderson wrote: > This looks like 6651434, which will be fixed in build NV82. It can't be. 6651434 is about idmap(1M), not idmapd. Can you send pstack output? Or can you send us the core dump? Thanks, Nico -- _______________________________________________ storage-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/storage-discuss
