[Bugs] [Bug 1151384] Rebalance fails to complete - stale file handles after 202, 908 files

bugzilla at redhat.com bugzilla at redhat.com
Fri Oct 17 07:34:19 UTC 2014


https://bugzilla.redhat.com/show_bug.cgi?id=1151384



--- Comment #2 from Raghavendra G <rgowdapp at redhat.com> ---
One possible reason for ESTALE errors is that the files/directories are deleted
after readdir is done. Brick processes return ESTALE in two scenarios:

1. during named-lookup when parent cannot be found.
2. during nameless lookup when gfid cannot be resolved (the error might be
ENOENT/ESTALE - need to confirm by reading code).
3. during any other fop when gfid passed with fop cannot be resolved.

If parallel I/O operations are done when rebalance is running, occurrence of
these logs is not an error-condition. However, we can fine-tune the behaviour
of rebalance when it sees these errors (like skipping that directory/file
instead of aborting rebalance altogether etc).

regards,
Raghavendra.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug https://bugzilla.redhat.com/token.cgi?t=DNORpM6fmK&a=cc_unsubscribe


More information about the Bugs mailing list