Date: Tue, 04 May 1999 14:11:02 -0500 From: Jacques Vidrine <n@nectar.com> To: Greg Lehey <grog@lemis.com> Cc: freebsd-questions@freebsd.org Subject: Re: vinum-- resizing disks Message-ID: <19990504191107.651582072@spawn.nectar.com> In-Reply-To: <19990504155450.U10134@freebie.lemis.com> References: <19990501210026.357911DC5@spawn.nectar.com> <19990502120821.W80561@freebie.lemis.com> <19990502070142.2AEF11DD7@spawn.nectar.com> <19990502163630.K80561@freebie.lemis.com> <19990502150018.1BA3D1DEB@spawn.nectar.com> <19990504155450.U10134@freebie.lemis.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 4 May 1999 at 15:54, Greg Lehey <grog@lemis.com> wrote: [snip] > Were you able to get a dump? That would have interested me. > > Obviously I was wrong in thinking that Vinum would automatically > detect what was going on. It only half did. I'll have to take a look > at the code and fix it, but it won't be in the next week. > > I'll send you private mail about how to fix the situation. Unfortunately, it went from bad to worse from there. I've since had to blow away my machine and restore from backup. Here are some of the situations that I recevied panics or the system hung. This is just anecdotal -- I did not have a kernel with debugging compiled in, so I didn't collect any useful information. I hope that later I'll be able to setup a scratch machine to reproduce them. * I created a large volume, and shortly after the system hung during newfs. * I removed that volume and tried recreating it, and the system hung. * I ``resetconfig'', rebuilt all of my volumes, and restored my data. The system hung when I did the following: # mount /dev/vinum/rel /rel # cd /rel # umount -f /rel # fstat . No such file or directory: . # fstat .. [hung] [several file system trashed, presumbably because I had a lot of activity going on] * I ``resetconfig'', rebuild all of my volumes, and restore my data. I notice that I never initialized the second plex on any of my mirrored volumes (the second plex status was ``initializing'', and the subdisks of that plex were ``empty''), so I attempt a few things (``vinum start'', ``vinum init'') to try to get the second plex sync'd, and fail miserbly, destroying the filesystem in the process. I remove and recreate the volume, and notice that the sizes listed by ``vinum ld'' are again incorrect. * I ``resetconfig'', and rebuild all of my volumes. Before newfs'ing of my mirrored volumes (usr), I do ``vinum init usr.p1''. This seems to do the right thing, so I newfs it and restore my data. I decide to do a little more testing by detaching and reattaching the plex. I do ``vinum detach usr.p1'', which seems to work fine. I can't fsck or mount the volume afterwards, though (errors indicate that the volume doesn't contain a recognizable UFS). So I attempt to re-attach the plex using ``vinum attach ex-usr.p1 usr'' but get only ``Invalid argument'' That's stuff I remember, and I list it only in case some of the symptoms are recognizable to readers of this list. After I get some other work done, I'm going to try to reproduce some of these things and trap some debugging info. I shall also try a more current version of vinum (this is all 3.1-STABLE circa April 29). Later, Jacques Vidrine / n@nectar.com / nectar@FreeBSD.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990504191107.651582072>