https://bugzilla.novell.com/show_bug.cgi?id=892847 https://bugzilla.novell.com/show_bug.cgi?id=892847#c0 Summary: [NTAP CQ599574] iscsi connections reconnect repeatedly after path failures Classification: openSUSE Product: openSUSE 13.1 Version: Final Platform: x86-64 OS/Version: SLES 11 Status: NEW Severity: Normal Priority: P5 - None Component: Basesystem AssignedTo: bnc-team-screening@forge.provo.novell.com ReportedBy: garrett.marks@netapp.com QAContact: qa-bugs@suse.de Found By: --- Blocker: --- User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:31.0) Gecko/20100101 Firefox/31.0 I'm testing iSER connections from SLES 11 SP3 to a NetApp Eseries iSER target. Often after a controller is reset or paths otherwise fail for some reason we see that one or more connections start repeatedly timing out and reconnecting immediately. This repeats for extended periods of time. On the host we see messages like the following. ug 20 15:54:39 ictb-cumberland kernel: [83473.414815] connection4:0: ping timeout of 5 secs expired, recv timeout 5, las t rx 4315799503, last ping 4315800753, now 4315802004 Aug 20 15:54:39 ictb-cumberland kernel: [83473.414832] connection4:0: detected conn error (1011) Aug 20 15:54:39 ictb-cumberland iscsid: Kernel reported iSCSI connection 4:0 error (1011 - ISCSI_ERR_CONN_FAILED: iSCSI co nnection failed) state (3) Aug 20 15:54:42 ictb-cumberland iscsid: connection4:0 is operational after recovery (1 attempts) Aug 20 15:55:06 ictb-cumberland kernel: [83500.337301] connection4:0: ping timeout of 5 secs expired, recv timeout 5, last rx 4315806244, last ping 4315807496, now 4315808748 Aug 20 15:55:06 ictb-cumberland kernel: [83500.337319] connection4:0: detected conn error (1011) Aug 20 15:55:06 ictb-cumberland iscsid: Kernel reported iSCSI connection 4:0 error (1011 - ISCSI_ERR_CONN_FAILED: iSCSI connection failed) state (3) Aug 20 15:55:09 ictb-cumberland iscsid: connection4:0 is operational after recovery (1 attempts) Aug 20 15:55:23 ictb-cumberland kernel: [83516.720811] connection4:0: ping timeout of 5 secs expired, recv timeout 5, last rx 4315810347, last ping 4315811600, now 4315812852 Aug 20 15:55:23 ictb-cumberland kernel: [83516.720827] connection4:0: detected conn error (1011) Aug 20 15:55:23 ictb-cumberland iscsid: Kernel reported iSCSI connection 4:0 error (1011 - ISCSI_ERR_CONN_FAILED: iSCSI connection failed) state (3) Aug 20 15:55:26 ictb-cumberland iscsid: connection4:0 is operational after recovery (1 attempts) It appears that the connection resets occur only when IO is running. If IO is not running on the host we haven't seen these connection resets. We've seen this with both mlx4 (connectx-3) and mlx5 (connect-ib) adapters. We're using the Mellanox OFED in this scenario. Reproducible: Always Steps to Reproduce: 1. Install Mellanox OFED 2.2-1.0.1 on a SLES 11 SP3 host 2. Configure the host to use iSER and map storage to the host from an iSER target. 3. Configure DMMP and run IO to the disks 4. Fail some of the paths to the storage. Actual Results: Some of the iscsi connections will start to disconnect and reconnect repeatedly. Expected Results: The iscsi connections should operate normally without disconnecting and reconnecting. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.