All I/O requests to this LV

[zxcv]

Hi guys,

we have hpux 11.23 with eva 6400 and rsm software..

we daily take 16 snapclones out of which today m gettng this error in my server syslog;

Nov  1 13:15:19 vishnu vmunix:          the PV becomes available (or a timeout is specified for the LV).
Nov  1 13:15:23 vishnu vmunix: LVM: NOTICE: VG 64 0x220000: LV 1: All I/O requests to this LV that were
Nov  1 13:15:23 vishnu vmunix:          waiting indefinitely for an unavailable PV have now completed.
Nov  1 13:16:20 vishnu vmunix: LVM: WARNING: VG 64 0x220000: LV 1: Some I/O requests to this LV are waiting
Nov  1 13:16:20 vishnu vmunix:  indefinitely for an unavailable PV. These requests will be queued until
Nov  1 13:16:20 vishnu vmunix:          the PV becomes available (or a timeout is specified for the LV).
Nov  1 13:16:23 vishnu vmunix: LVM: NOTICE: VG 64 0x220000: LV 1: All I/O requests to this LV that were
Nov  1 13:16:23 vishnu vmunix:          waiting indefinitely for an unavailable PV have now completed.
Nov  1 13:17:03 vishnu cmdisklockd[3361]: Still trying to inquire cluster lock disk /dev/dsk/c2t0d1
Nov  1 13:18:30 vishnu cmdisklockd[3361]: Still trying to inquire cluster lock disk /dev/dsk/c2t0d1
Nov  1 13:19:25 vishnu vmunix: LVM: WARNING: VG 64 0x220000: LV 1: Some I/O requests to this LV are waiting
Nov  1 13:19:25 vishnu vmunix:  indefinitely for an unavailable PV. These requests will be queued until
Nov  1 13:19:25 vishnu vmunix:          the PV becomes available (or a timeout is specified for the LV).
Nov  1 13:19:26 vishnu vmunix: LVM: NOTICE: VG 64 0x220000: LV 1: All I/O requests to this LV that were
Nov  1 13:19:26 vishnu vmunix:          waiting indefinitely for an unavailable PV have now completed.

As a result of which my that particular vg has been choked and my whole applcn is hanegd.

please suggest what is to be done.

[Ivan Lyanguzov]

zxcv,

There could be many reasons for this behavior. I would recommend to open support case. Possible causes may include:
1. Some issue in SAN (failed switch, SFP, inter-switch link or similar).
2. Issues on the EVA side (failed controller, controller port or similar).
3. Incorrect HP-UX configuration (multi-path/path balancing, PV timeouts or other).

So, it is best to contact HP support.

Ivan.