Date: Thu, 4 Dec 2014 17:51:11 +0100 From: Walter Hop <freebsd@spam.lifeforms.nl> To: freebsd-stable@FreeBSD.org Subject: Re: System hang on shutdown when running freebsd-update Message-ID: <20FF25C9-B2D3-490C-BD01-6F834017BDA1@spam.lifeforms.nl> In-Reply-To: <5B600B90-9967-4031-AB9B-40ADDBE56CAF@spam.lifeforms.nl> References: <2B4EEDA7-C3D9-465A-B0C9-B5728D438077@spam.lifeforms.nl> <5B600B90-9967-4031-AB9B-40ADDBE56CAF@spam.lifeforms.nl>
next in thread | previous in thread | raw e-mail | index | archive | help
Another update. I narrowed down the issue due to /sbin/init being replaced. This might be the magic that freebsd-update does to make the crash happen. (Although there might be more situations that cause a hang.) I can completely reliably trigger the hang on a default 10.1-RELEASE install on UFS2 in VMware Fusion with the following procedure: # chflags noschg /sbin/init # cp -Rp /sbin/init /sbin/init2 # rm -f /sbin/init # mv /sbin/init2 /sbin/init # chflags schg /sbin/init # reboot => Hang after "All buffers synced." This looks useful because we don’t have to do a full freebsd-update to get the hang now. I’d be interested to see if others can reproduce, because for me in VMware it’s 100% of the times. It doesn’t happen on 10.0 kernel, also not on 10.1 ZFS, and not on 10.1 UFS2 with softupdates disabled. We updated 15 machines to 10.1 with a modified upgrade procedure (first disable softupdates, then upgrade to 10.1, then re-enable softupdates). Without softupdates there’s no lockup. So: 10.1 + UFS2 + softupdates + replacing /sbin/init = hang+100% CPU on next reboot/root unmount/root remount as readonly. I don’t know if we can research more, I wonder what would happen on CURRENT but I don’t have time to build right now... Cheers, WH > On 29 Nov 2014, at 13:17, Walter Hop <freebsd@spam.lifeforms.nl> wrote: > > I’m revisiting this issue, since unfortunately I still have it more often than not when upgrading to 10.1-RELEASE. > > As Kevin Oberman suspected earlier in the thread, the issue seems to lie in unmounting. The same hang occurs when dropping to single user mode and trying to re-mount root as readonly. > > I’ve also had another unmount issue after upgrading to 10.1-RELEASE: > > All buffers synced. > softdep_waitidle: Failed to flush worklist for 0xfffff800027b4330 > unmount of / failed (BUSY) > > I’ve created a PR with the information I have: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=195458 > > With the EOL date of FreeBSD 10.0 on the horizon, it’s making me a little skittish. > Any ideas of experiments I can do to get more info out of a problematic box, or other options to take? > > Thanks! > WH > >> On 28 Oct 2014, at 23:09, Walter Hop <freebsd@spam.lifeforms.nl> wrote: >> >> I noticed this same hang after upgrading from 10.0-RELEASE to 10.1-RC3 in a VM running under VMware Fusion, so the problem appears still present. >> >> I could only make it happen in the single uptime just after the system was freebsd-updated from FreeBSD 10.0 to 10.1-RC3. >> >> Here is a screenshot: http://lf.ms/wait-for-reboot.png >> >> It did not make any progress after 2 hours of waiting. When restarting the VM, the disk was dirty. >> >> Some interesting facts: >> - Note "swapoff: /dev/da0p2: Cannot allocate memory" in the screenshot which might pose a clue. I haven’t seen this normally. >> - FreeBSD does respond to ping while it is busy, so it is not a complete "freeze". >> - The VM is at 100% CPU while this is going on. > > -- > Walter Hop | PGP key: https://lifeforms.nl/pgp > > _______________________________________________ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "freebsd-stable-unsubscribe@freebsd.org" -- Walter Hop | PGP key: https://lifeforms.nl/pgp
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20FF25C9-B2D3-490C-BD01-6F834017BDA1>