Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On Thu, Jun 08, 2006 at 09:16:49AM -0400, lu.guanqing at connexim.ca wrote: > Hi, > > We have run drbd between two nodes with RHEL3.3 OS since last > November. Yesterday, we had input/output error: nobody, even not root, > could write into the filesystem. The problem was corrected by > shutdowning the drbd service and filesystem checking. When I did the > fsck, there were a lot of inode errors. I remember that we deleted a lot > of log files in the drbd filesystem two weeks ago. > > My question are: > 1. Is deleting a lot of files the cause of input/output error in the > drbd filesystem? hard to be certain, given the vagueness of you error message, but it seems unlikely to me: see next answer too. Are you sure it is not an error on the underlying hardware (disk, array or whatever) ? Perhaps then the kernel might then remount the filesystem read-only ? or perhaps drbd forces the block device to read-only with some error-conditions/config ? I would be wary of assuming that the process you outline would fix your problem, rather than just making it go away until the next time you hit that bad block. > 2. is it a regular process to do an fsck after deleting a lot of files > in the drbd filesystem? drbd is not a filesystem, it is a block device. IIRC, there are circumstances under which you would want to fsck the filesystem on top of a drbd, but I'm not aware of anything that is related soley to the volume of traffic. Hope this helps. Regards, Paddy -- Perl 6 will give you the big knob. -- Larry Wall