Actually, it might have worked better if the PG had stayed down while running cephfs-data-scan, as it could only then get file structure from metadata pool and not touch each file/link in data pool? This would at least properly have given the list of files in (only) the affected PG? //Jesper ________________________________ Fra: Jesper Lykkegaard Karlsen <jelka@xxxxxxxxx> Sendt: 19. august 2022 22:49 Til: Patrick Donnelly <pdonnell@xxxxxxxxxx> Cc: ceph-users@xxxxxxx <ceph-users@xxxxxxx> Emne: Re: Potential bug in cephfs-data-scan? ________________________________ Fra: Patrick Donnelly <pdonnell@xxxxxxxxxx> Sendt: 19. august 2022 16:16 Til: Jesper Lykkegaard Karlsen <jelka@xxxxxxxxx> Cc: ceph-users@xxxxxxx <ceph-users@xxxxxxx> Emne: Re: Potential bug in cephfs-data-scan? On Fri, Aug 19, 2022 at 5:02 AM Jesper Lykkegaard Karlsen <jelka@xxxxxxxxx> wrote: >> > >Hi, >> >> I have recently been scanning the files in a PG with "cephfs-data-scan pg_files ...". >Why? I had an incident where a PG that went down+incomplete after some OSD crashed + heavy load + ongoing snap trimming. Got it back up again with object store tool by marking complete. Then I wanted to show possible affected files with cephfs-data-scan in the unfortunate PG, so I could recover potential loss from backup. >> Although, after a long time the scan was still running and the list of files consumed 44 GB, I stopped it, as something obviously was very wrong. >> >> It turns out some users had symlinks that looped and even a user had a symlink to "/". >Symlinks are not stored in the data pool. This should be irrelevant. Okay, it may be a case of me "holding it wrong", but I do see "cephfs-data-scan pg_files" trying to follow any global or local symlink in the file structure, which leads to many more files registrered than possibly could be in that PG and even endless loops in some cases. If the symlinks are not stored in data pool, how can cephfs-data-scan then follow the link? And how do I get "cephfs-data-scan" to just show the symlinks as links and not follow them up or down in directory structure? Best, Jesper _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |