lost connection to hsv-110 controller on eva 5000

[Paul Coviello]

valid point about drives... vcs 3 maybe? not sure

[Sebastian.Koehler]

If drives are red/down there are several options...

1. Reseat the drive / wait 2min before reinstall / you can do that
2. Drives have been failed and need to be replaced / support may help
3. Drives are on DSL (drive suspect list) / service can remove them

Regards,
Sebastian

[Paul Coviello]

nope everything is green...nothing red

[Torsten.]

The question is not what is green or red, but what LED is off and should be green (rear side)?

Based on experience there are 2 likely reasons for the problem:
1) a bad component (disk) prevents communication between disks and controllers by reseting the loop again and again

2) a corrupted (default) diskgroup

Both are hard to identify. So let HP analyze the logs and keep your backup handy - hopefully you won't need it.

[Paul Coviello]

everything that I could see is on in the back.  got quote for a time and labor for hp to send a hw tech out.  so what is the worse that I can do by starting one of the controllers?

thanks

Paul

[Sebastian.Koehler]

If you only touch the controllers, the worst thing would be loss of controller logs that would make root cause analysis impossible. If you have old code, they will not perform any RCA and force you to the latest VCS anyway.

If you decide to touch the controllers, remove power from both and bring them up together again. Just keep an eye on the OCP and DO NOT ENTER ANY WWN/CHECKSUM if prompted. The system has to find the metadata on the quorum disks spread among the disk groups.

Regards,
Sebastian

[Paul Coviello]

just tried to go through menu options on front panel and nothing changes. I did notice the lights for the switch ports on the back of the controllers are not on. the ones on the switch are.

so it would be hit power switch on back and then back on? for both at one time?

thanks

Paul

[Sebastian.Koehler]

Yes, turn them off and wait 2min before turn them on again. Cycle both controllers to make sure the other controller does not lock metadata access. You will see both controllers go trough their TEs then and hopefully boot up correctly.

Regards,
Sebastian

[Paul Coviello]

I know silly question at this point since there is no clue as to the state of the disks, but since this is a cluster of 2 and the node attached to this SAN has their system disk in a mount verify state but still up... so what should happen to that disk  and cluster transition? (VMS)  when I reboot the controllers?

thanks

Paul

[Torsten.]

I'm sure none of your servers can currently access this EVA, because this array is not up and running.