On Mon, Mar 20, 2023 at 06:33:49PM +0000, Andrew M.A. Cater wrote: > On Mon, Mar 20, 2023 at 07:01:22PM +0100, john doe wrote: > > Debians, > > > > I'm seeing some alarming MSGs (E.G: ' blk_update_request: I/O error, dev > > sda, sector N') in my server's log. > > As it's read-only, I can not install smartmontools to investigate further.
[...] > Run, don't walk, to another computer and order new disks and shut this one > down immediately. If the file system has gone read only because of errors, > you need to shut it down NOW, maybe, and hope that you can recover data when > it boots up. Basically yes, but that'll depend on what is more valuable. If it's the data, that's spot-on. The longer the thing runs, the higher the risk of loss. If it's the uptime, then don't touch it and don't shut it down for as long as possible. The data would be less valuable if you have backups which will be "good enough" to resucitate your set up from. > Good luck, but it's probably too late. Crossing fingers. Cheers -- t
signature.asc
Description: PGP signature
