kernel: mptscsih: ioc0: WARNING - Issuing Reset from mptscsih_IssueTaskMgmt!! doorbell=0x24000000

Kamal Luthra · 06-04-2020, 01:56 PM

Hi All,
I am observing below messages in /var/log/messages from past few weeks on different servers and when checking SAN logs nothing found suspicious. I am using 3par SAN storage hardware "BL460c Gen8" mapped to CENT OS version release 7.5.1804 (Core) running on VmWare ESXi Version as 6.0.0, 3620759.

Jun 4 00:35:22 WES-001-US-1023 kernel: mptscsih: ioc0: attempting task abort! (sc=ffff902278e43500)
Jun 4 00:35:22 WES-001-US-1023 kernel: sd 0:0:0:0: [sda] CDB: Write(10) 2a 00 03 70 4f c0 00 04 00 00
Jun 4 00:36:41 WES-001-US-1023 kernel: mptscsih: ioc0: WARNING - Issuing Reset from mptscsih_IssueTaskMgmt!! doorbell=0x24000000
Jun 4 00:36:41 WES-001-US-1023 kernel: mptscsih: ioc0: task abort: SUCCESS (rv=2002) (sc=ffff902278e43500)
Jun 4 00:36:41 WES-001-US-1023 kernel: mptscsih: ioc0: attempting task abort! (sc=ffff90232a721dc0)
Jun 4 00:36:41 WES-001-US-1023 kernel: sd 0:0:0:0: [sda] CDB: Write(10) 2a 00 03 70 5f c0 00 04 00 00
Jun 4 00:36:41 WES-001-US-1023 kernel: mptscsih: ioc0: task abort: SUCCESS (rv=2002) (sc=ffff90232a721dc0)
Jun 4 00:36:41 WES-001-US-1023 kernel: mptscsih: ioc0: attempting task abort! (sc=ffff90232a720c40)
Jun 4 00:36:41 WES-001-US-1023 kernel: sd 0:0:0:0: [sda] CDB: Write(10) 2a 00 03 70 77 c0 00 04 00 00
Jun 4 00:36:41 WES-001-US-1023 kernel: mptscsih: ioc0: task abort: SUCCESS (rv=2002) (sc=ffff90232a720c40)
Jun 4 00:36:41 WES-001-US-1023 kernel: mptscsih: ioc0: attempting task abort! (sc=ffff902278e43c00)
Jun 4 00:36:41 WES-001-US-1023 kernel: sd 0:0:0:0: [sda] CDB: Write(10) 2a 00 03 70 87 c0 00 04 00 00
Jun 4 00:36:41 WES-001-US-1023 kernel: mptscsih: ioc0: task abort: SUCCESS (rv=2002) (sc=ffff902278e43c00)
Jun 4 00:36:41 WES-001-US-1023 kernel: mptscsih: ioc0: attempting task abort! (sc=ffff902278e43a40)
Jun 4 00:36:41 WES-001-US-1023 kernel: sd 0:0:0:0: [sda] CDB: Write(10) 2a 00 03 70 8b c0 00 04 00 00

Your urgent assistance is appreciated.

Regards,
Kamal

smallpond · 06-12-2020, 11:18 AM

My guess is that the writes are timing out, and the system is sending abort to do recovery and retry. Check for errors on the storage or the VM host which might indicate a hardware problem. If there aren't any, then you could try increasing the command timeout on the VM in /sys/block/sda/device/timeout which is probably 30 seconds by default. How busy is the storage?