Details
-
Improvement
-
Resolution: Fixed
-
Critical
-
Lustre 2.9.0
-
9223372036854775807
Description
There at occasions where the OI file can become corrupted (e.g. double mount of a filesystem), or otherwise need to be rebuilt (e.g. upgrade of normal ZFS filesystem to Lustre MDT in the future). It would be useful to be able to verify and/or rebuild the ZFS OI files during dnode iteration.
Attachments
Issue Links
- is related to
-
LU-7797 Can't mount zpools after OSS restart
- Resolved
-
LU-3569 Use real OST index as ostid_to_fid() parameter instead of always "0"
- Resolved
-
LU-5855 sanity-lfsck does not work under ZFS-based DNE mode
- Resolved
-
LU-7582 Manual delete of oi directory in ZFS backed server causes panic
- Resolved
-
LU-8521 ZFS OST is unwritable
- Closed
-
LUDOC-161 document backup/restore process for ZFS backing filesystems
- Resolved
-
LU-8620 Get rid of mystery object '???<object#138>' from the output of zdb
- Closed
1.
|
OI compatibility between ldiskfs and ZFS | Resolved | nasf (Inactive) | |
2.
|
Handle FID-in-dirent | Resolved | nasf (Inactive) | |
3.
|
Block size alignment issues | Resolved | nasf (Inactive) | |
4.
|
Handle PFID EA in LMA | Resolved | nasf (Inactive) | |
5.
|
Agent object for cross-MDTs reference | Resolved | nasf (Inactive) | |
6.
|
Agent entry for cross-MDTs reference | Resolved | nasf (Inactive) | |
7.
|
Index object on-disk layout compatibility | Resolved | nasf (Inactive) |