This morning, a busy NetBSD-8.0/amd64 server greeted me with an unpleasant panic on reboot: ffs_snapshot_mount: xxx already on list
I booted in single user mode and ran fsck. After cleaning up things, the filesystem would mount read-only, but mounting read-write, with or without -o log, would raise the same panic. There is no mount option to disable snapshots, hence I tried to build a custom kernel with options FFS_NO_SNAPSHOT to run without snapshot support and get this filesystem to mount. No luck, even with so called FFS_NO_SNAPSHOT, it would still crash in snapshot-related ffs_snapshot_mount. I patched ffs_snapshot_mount to #ifndef FFS_NO_SNAPSHOT, the whole function, rebuilt a kernel and booted on that. With that, I was able to mount read-only. I then removed the obsolete snapshot backend files from the filesystem and rebooted on GENERIC. No panic anymore, it only complains at mount time: ffs_snapshot_mount: vget failed 2 ffs_snapshot_mount: vget failed 2 ffs_snapshot_mount: vget failed 2 Beside the problem that FFS_NO_SNAPSHOT does not really disable snapshot code, I think we should have a nosnapshot or nofss mount option to handle such a scenario. Anyone has opinion on this? -- Emmanuel Dreyfus http://hcpnet.free.fr/pubz m...@netbsd.org