[Bugs] [Bug 1449329] When either killing or restarting a brick with performance.stat-prefetch on , stat sometimes returns a bad st_size value.

[bugzilla at redhat.com]

https://bugzilla.redhat.com/show_bug.cgi?id=1449329

Ravishankar N <ravishankar at redhat.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
              Flags|                            |needinfo?(miklos.fokin at appe
                   |                            |artv.com)



--- Comment #6 from Ravishankar N <ravishankar at redhat.com> ---
So I ran it for about 90 minutes plus  but could not hit the issue.

Can you add the following log to see if afr_fsync_cbk is indeed the one that is
sending a short inode size when you run your test?
--------------------------
diff --git a/xlators/cluster/afr/src/afr-common.c
b/xlators/cluster/afr/src/afr-common.c
index 34d9e56..cd03056 100644
--- a/xlators/cluster/afr/src/afr-common.c
+++ b/xlators/cluster/afr/src/afr-common.c
@@ -3371,6 +3371,7 @@ afr_fsync_cbk (call_frame_t *frame, void *cookie,
xlator_t *this,
                         if (i == read_subvol)
                                 break;
                 }
+                gf_log (this->name, GF_LOG_INFO, "Unwinding with a size of %ld
bytes from brick:%d", local->cont.inode_wfop.postbuf.ia_size, read_subvol);

                /* Make a stub out of the frame, and register it
                   with the waking up post-op. When the call-stub resumes,

--------------------------

-- 
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=xV93w7Suw0&a=cc_unsubscribe