Please don't reply to lustre-devel. Instead, comment in Bugzilla by using the
following link:
https://bugzilla.lustre.org/show_bug.cgi?id=11330
What |Removed |Added
----------------------------------------------------------------------------
Attachment #9198|review?([EMAIL PROTECTED])|review+
Flag| |
(From update of attachment 9198)
Looks good.
Few minor comments.
>@@ -695,8 +698,21 @@ static int ost_brw_read(struct ptlrpc_re
>+ /* fix the bug11330 */
This comment is too cryptic.
Better to say something like 'Check if getting the lock took more time than
client was willing to wait'
>+ do_gettimeofday(&present_time);
>+ timediff = cfs_timeval_sub(&present_time,
>&req->rq_arrival_time,NULL);
>+ if (timediff / 1000000 > (long)obd_timeout) {
>+ no_reply = 1;
>+ CERROR("Dropping timed-out opc %d request from %s"
>+ ": %ld seconds old\n", req->rq_reqmsg->opc,
>+ libcfs_id2str(req->rq_peer),
>+ timediff / 1000000);
The error message here should be a bit changed too, I think.
Something like 'Dropping opc ... because taking the lock took too long' (and
print resource for the lock too,
I think, now that we actully have it)
Same in ost_brw_write.
Also we probably need a test and FAIL_LOC injected to actually test this
codepath (and recovert-small.sh test too?)
_______________________________________________
Lustre-devel mailing list
[email protected]
https://mail.clusterfs.com/mailman/listinfo/lustre-devel
