If redo fails, you can set a tunable:
vfs.hammer.skip_redo=<value>
0 Run redo recovery normally and fail to mount in the case of
error
(default).
1 Run redo recovery but continue mounting if an error appears.
2 Completely bypass redo recovery.
I assume you'd want 2, and then immediately copy your data to a filesystem
you can trust on another machine. I don't know what's up with that
machine, but take no chances.
There is a 'hammer recover' command (see man 8 hammer), to use as a
last-ditch effort. I've never tried it.
On Sat, Oct 10, 2015 at 7:29 PM, Konrad Neuwirth <[email protected]> wrote:
> Dear reader,
>
> our primary server just rebooted and has somehow shot itself into the
> hammer filesystem while doing so.
>
> It can’t mount hammer because undo/redo recovery doesn’t work.
>
> I can only access the machine via its console card, so I can’t copy &
> paste text contents, but what I see is:
>
>
> So I can mount read-only, but it doesn’t work so well for read-write.
>
> I need the data on that filesystem. If one or two files are a problem,
> that’s not tough.
>
> What could I do?
>
> Thank you,
> Konrad
>
