Sadly, it's probably a known bug that will be fixed in 1.6.1, coming soon (as soon as buildbot catches up).
On Wed, Oct 5, 2011 at 5:43 PM, John Tang Boyland <boyl...@pabst.cs.uwm.edu> wrote: > I've had a problem with the new DA file server. It lost an entire directory > and the file server crashed. The symptom that something was going bad was > > % touch 176 > touch: cannot create 176: File too large > pabst.cs 297 % fs lq . > > Volume Name Quota Used %Used Partition > fa11.cs732 100000 64776 65% 1% > > This happened before, the directory was impossible to use. > (That is, one could not create any files in it.) > I created a new directory and abandoned the old directory, > but then a few weeks later, the problem reocurred. In both cases, > the FileLog has entries such as: > Wed Oct 5 16:08:38 2011 ReallyRead(): read failed device 0 inode > 7069583278211 errno 5 > > But this time, when I tried to rename (mv) the directory > which had the problem, the fileserver crashed: > Fatal Rx error: assertion failed: Create(&filedir, "..", NewDirFid) == 0, > file: ../viced/afsfileprocs.c, line: 4215 > > When it came back up, the directory where files could not be created > was empty. (But the directory in the backup volume still had > all the files, fortunately.) > > The /vicep* partitions are ext4. Maybe that's the problem? > > I know 1.6.0rc2 is old, but it's the latest version available > with the Scientific Linux 6 "yum update". If it's a known bug, > I'll just have to wait. (Or if "yum update openafs-server" > isn't the command I'm supposed to use, let me know.) > > Best regards, > John > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > -- Derrick _______________________________________________ OpenAFS-info mailing list OpenAFS-info@openafs.org https://lists.openafs.org/mailman/listinfo/openafs-info
