Port timeout errors

[Sean Carolan]

We are starting to see this error occurring more frequently. I analyzed the last ten instances and found that the enclosure was always the same, however the drive bays were different. Command View EVA shows healthy status for all hardware and disk groups.

Is this a cause for concern?

SNMP Alert:
An exchange sent to a physical disk drive or another HSV200 controller via the mirror port or a Fibre Channel port has timed out.
Corrective action code: 47

Event Log:
device - UUID of the physical disk drive, HSV200 controller, or Drive Enclosure Environmental Monitoring Unit associated with the event: 2000-001d-3812-225c-0000-0000-0000-0000
cerp_id - HSV200 controller enclosure rear panel Fibre Channel port attached to the physical disk drive or mirror port: DP-1A
al_pa - AL_PA of the physical disk drive or the mirror port: 92 [0x5c]
dencl_num - Enclosure where the physical disk drive is located: 3 [0x03]
port - HSV200 controller internal Fibre Channel port number attached to the physical disk drive or mirror port: 0 [0x00]
rack_num - Rack where the physical disk drive is located: 0 [0x00]
bay - Enclosure bay where the physical disk drive is located: 7 [0x07]
fed_class - Fibre Channel Exchange Descriptor class: 33554480 [0x02000030]
num_times - Number of occurrences of the error.: 1 [0x01]

[Víctor Cespón]

Get the log of this EVA checked. The number of timeouts on each disk need to be counted, and see if there are other events.

The disks on enclosure 3 are 1 TB FATA with firmware below HP06?

[Sean Carolan]

Yes. I'll open a support ticket with HP tomorrow morning. One of the disks shows "27 occurrences", but oddly it says 27 every time the error occurs. The others just say 1.

[Kites BM]

>>>>>>>SNMP Alert:
An exchange sent to a physical disk drive or another HSV200 controller via the mirror port or a Fibre Channel port has timed out.
Corrective action code: 47

This is a bug that needs immediate fix.

[Sean Carolan]

It's a bug? How do you know?

[Clint Placette_2]

Lets not get ahead of our selves. Unless product engineering or 2 lvl support said it was a bug then lets not assume.

You could have a drive going bad that may have been called out then it corrected it's issue. If this disk continues to fail there is a algorithm that will mark the drive bad and remove it from the the DG. The count is around ~9999 before it is remove into a Drive suspect list.

Sometime the problem is not the disk that is reporting an issue but maybe another disk in the same enclosue that is causing the issue. Or could be a combination multiple disk issue.

Depending on your firmware version the EVA may/maynot know how to handle the error. You need to open a case with HP and send them the parse log file so they can analyse indepth the issue and they most likely will find a pattern that can point directly to the offinding hardware.