staging: lustre: mgc: IR log failure should not stop mount
authorwang di <di.wang@intel.com>
Thu, 10 Nov 2016 17:30:39 +0000 (12:30 -0500)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Mon, 14 Nov 2016 15:02:55 +0000 (16:02 +0100)
If clients or other targets can not get IR config lock
or lock, the mount should continue, instead of failing.
Because timeout mechanism will handle the recovery anyway.

Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6906
Reviewed-on: http://review.whamcloud.com/15728
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/staging/lustre/lustre/mgc/mgc_request.c

index 2d6fdd0..e248f33 100644 (file)
@@ -1675,8 +1675,15 @@ restart:
 
        if (cld_is_recover(cld)) {
                rc = 0; /* this is not a fatal error for recover log */
-               if (rcl == 0)
+               if (!rcl) {
                        rc = mgc_process_recover_log(mgc, cld);
+                       if (rc) {
+                               CERROR("%s: recover log %s failed: rc = %d not fatal.\n",
+                                      mgc->obd_name, cld->cld_logname, rc);
+                               rc = 0;
+                               cld->cld_lostlock = 1;
+                       }
+               }
        } else {
                rc = mgc_process_cfg_log(mgc, cld, rcl != 0);
        }