Thanks!
-Eli
On Wed, Sep 30, 2009 at 1:36 PM, Eli Klein <elijah@xxxxxxxxx> wrote:
On Sep 30, 2009, at 1:24 PM, John Brier wrote:On 09/30/2009 03:16 PM, Eli Klein wrote:Here's the abbreviated output from multipath -v3:mpath1: pgfailback = 120 (config file default)mpath1: pgpolicy = multibus (config file default)mpath1: selector = round-robin 0 (internal default)
mpath1: features = 1 queue_if_no_path (internal default)mpath1: hwhandler = 0 (internal default)mpath1: rr_weight = 2 (config file default)
mpath1: minio = 10 (config file default)mpath1: no_path_retry = 1 (config file default)pg_timeout = NONE (internal default)
mpath1: set ACT_NOTHING (map unchanged)It's definitely picking up the settings correctly from multipath.conf,
but the paths are still reinstated immediately once they're reconnected(even if it's just for a split second).
What does /var/log/messages look like when this happens? Since you changed failback to 120 did you restart multipathd or restart the box?
John Brier
--
dm-devel mailing list
dm-devel@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/dm-develI've tried rebooting, restarting multipath (and iscsid), flushing multipath (-F), removing all the iscsi targets completely and then reinitializing the targets. Nothing changes the behavior. Here's output from /var/log/messages.. note the time between when the link becomes active (13:31:08) to when the path is reinstated (13:31:15):Sep 30 13:30:28 testbox kernel: connection7:0: ping timeout of 5 secs expired, recv timeout 5, last rx 4470061909, last ping 4470066909, now 4470071909Sep 30 13:30:28 testbox kernel: connection7:0: detected conn error (1011)Sep 30 13:30:28 testbox multipathd: sdf: readsector0 checker reports path is downSep 30 13:30:28 testbox multipathd: checker failed path 8:80 in map mpath0Sep 30 13:30:28 testbox multipathd: mpath0: remaining active paths: 1Sep 30 13:30:28 testbox kernel: device-mapper: multipath: Failing path 8:80.Sep 30 13:30:28 testbox iscsid: Kernel reported iSCSI connection 7:0 error (1011) state (3)Sep 30 13:30:29 testbox kernel: connection9:0: ping timeout of 5 secs expired, recv timeout 5, last rx 4470063232, last ping 4470068232, now 4470073232Sep 30 13:30:29 testbox kernel: connection9:0: detected conn error (1011)Sep 30 13:30:29 testbox multipathd: sdg: readsector0 checker reports path is downSep 30 13:30:29 testbox multipathd: checker failed path 8:96 in map mpath1Sep 30 13:30:29 testbox multipathd: mpath1: remaining active paths: 1Sep 30 13:30:29 testbox kernel: device-mapper: multipath: Failing path 8:96.Sep 30 13:30:30 testbox kernel: connection11:0: ping timeout of 5 secs expired, recv timeout 5, last rx 4470064373, last ping 4470069373, now 4470074373Sep 30 13:30:30 testbox kernel: connection11:0: detected conn error (1011)Sep 30 13:30:30 testbox iscsid: Kernel reported iSCSI connection 9:0 error (1011) state (3)Sep 30 13:30:30 testbox multipathd: sdh: readsector0 checker reports path is downSep 30 13:30:30 testbox multipathd: checker failed path 8:112 in map mpath2Sep 30 13:30:30 testbox kernel: device-mapper: multipath: Failing path 8:112.Sep 30 13:30:30 testbox multipathd: mpath2: remaining active paths: 1Sep 30 13:30:30 testbox multipathd: dm-4: add map (uevent)Sep 30 13:30:30 testbox multipathd: dm-4: devmap already registeredSep 30 13:30:30 testbox multipathd: dm-2: add map (uevent)Sep 30 13:30:30 testbox multipathd: dm-2: devmap already registeredSep 30 13:30:30 testbox multipathd: dm-3: add map (uevent)Sep 30 13:30:30 testbox multipathd: dm-3: devmap already registeredSep 30 13:30:31 testbox iscsid: Kernel reported iSCSI connection 11:0 error (1011) state (3)Sep 30 13:30:43 testbox kernel: session7: session recovery timed out after 15 secsSep 30 13:30:43 testbox multipathd: sdf: readsector0 checker reports path is downSep 30 13:30:44 testbox kernel: session9: session recovery timed out after 15 secsSep 30 13:30:44 testbox multipathd: sdg: readsector0 checker reports path is downSep 30 13:30:45 testbox kernel: session11: session recovery timed out after 15 secsSep 30 13:30:45 testbox kernel: sd 13:0:0:0: SCSI error: return code = 0x000f0000Sep 30 13:30:45 testbox kernel: end_request: I/O error, dev sdh, sector 407978223Sep 30 13:30:45 testbox multipathd: sdh: readsector0 checker reports path is downSep 30 13:30:55 testbox multipathd: sdf: readsector0 checker reports path is downSep 30 13:31:05 testbox multipathd: sdg: readsector0 checker reports path is downSep 30 13:31:05 testbox multipathd: sdh: readsector0 checker reports path is downSep 30 13:31:05 testbox multipathd: sdf: readsector0 checker reports path is downSep 30 13:31:05 testbox multipathd: sdg: readsector0 checker reports path is downSep 30 13:31:05 testbox multipathd: sdh: readsector0 checker reports path is downSep 30 13:31:08 testbox iscsid: connection9:0 is operational after recovery (3 attempts)Sep 30 13:31:08 testbox iscsid: connection11:0 is operational after recovery (3 attempts)Sep 30 13:31:11 testbox iscsid: connection7:0 is operational after recovery (3 attempts)Sep 30 13:31:15 testbox multipathd: sdf: readsector0 checker reports path is upSep 30 13:31:15 testbox multipathd: 8:80: reinstatedSep 30 13:31:15 testbox multipathd: mpath0: remaining active paths: 2Sep 30 13:31:15 testbox multipathd: sdg: readsector0 checker reports path is upSep 30 13:31:15 testbox multipathd: 8:96: reinstatedSep 30 13:31:15 testbox multipathd: mpath1: remaining active paths: 2Sep 30 13:31:15 testbox multipathd: sdh: readsector0 checker reports path is upSep 30 13:31:15 testbox multipathd: 8:112: reinstatedSep 30 13:31:15 testbox multipathd: mpath2: remaining active paths: 2Sep 30 13:31:15 testbox multipathd: dm-4: add map (uevent)Sep 30 13:31:15 testbox multipathd: dm-4: devmap already registeredSep 30 13:31:15 testbox multipathd: dm-2: add map (uevent)Sep 30 13:31:15 testbox multipathd: dm-2: devmap already registeredSep 30 13:31:15 testbox multipathd: dm-3: add map (uevent)Sep 30 13:31:15 testbox multipathd: dm-3: devmap already registered
-- dm-devel mailing list dm-devel@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/dm-devel
