Date: Fri, 03 Feb 2017 23:02:17 +0000 From: bugzilla-noreply@freebsd.org To: freebsd-fs@FreeBSD.org Subject: [Bug 216586] zfs panic: sa.sa_magic == 0x2F505A in zfs_space_delta_cb() Message-ID: <bug-216586-3630-742BFKQnLx@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-216586-3630@https.bugs.freebsd.org/bugzilla/>
index | next in thread | previous in thread | raw e-mail
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=216586 --- Comment #4 from johannes@jo-t.de --- Thanks Andriy. There are indeed lots of (very) old files there. I had an earlier snapshot from a few years ago (that refs all these files). Deleted all files in the fs, and rsync'd everything anew. No panic. I've also tried (first rolling back to the old snapshot with all old files present) rsync with either --xattrs present or not. Panics for both cases. But the time until panic varies a lot. So looks like a race condition. Question is now, is there bogus data (old format that wasn't properly upgraded on-the-fly) with the correct checksum on the disk, and the current code trips over it. Or is the race condition still present when it tries to on-the-fly-upgrade old files? -- You are receiving this mail because: You are the assignee for the bug.home | help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-216586-3630-742BFKQnLx>