MSA 1000 enclosure replacement / WWN Numbers

[Christian Bieletzki]

Hi,

I must exchange the MSA1000 enclosure due to a hardware problem with the backplane. in which component are the WWN numbers for the controllers saved? Or do I not have to update the zoning configuration? I'm just wondering which component holds this. The controllers are hot swappable, so they should not hold it.

Christian

[Uwe Zessin · Accepted solution]

The WWN is stored in a chip on the backplane. I don't know if it can be swapped between good and bad backplanes - I am afraid not as I guess it holds more data like serial numbers, but let us know if you find out.

[Christian Bieletzki]

Thank you for that information.
We had to change the WWN but I was really interested where this is stored because almost all components are hot swappable.

Best regards
Christian

By the way: we changed the enclosure because of the following error message:

Array Controller \Device\FibreArray0, HBA Slot 3, Chassis: GB204110TP is no longer redundant because the inter-controller link has failed. The standby array controller has been marked as failed.

This occured sometimes during normal operation without any reason. Do you possibly know if this error message appears as well if you trigger a manual "move to other controller"

Christian

[Uwe Zessin]

Unfortunately (or better, luckily) I have never been able to catch the log message from a controller failure, so I cannot tell, sorry. And I have never seen this message while working with MSA1000s - I even checked some old logs I had stored away.

[Christian Bieletzki]

Are you working with emulex or qlogic?

This sound not very good to me.
What else could that problem be related to?

We already changed

Enclosure
FC controllers
Cache modules

If you start a server and the agents are polling the status of the MSA1000, do they get all messages stored in the MSA since the last poll or just the ones they can gather during uptime...

We pulled out one controller after the other after we exchanged the parts which justifies the error message...but this was during server downtime.

Best regards
Christian

[Uwe Zessin]

We're selling Qlogic by default, but it looks like there is a misunderstanding:

I thought you were talking about CLI messages. When I am working with the MSA1000 it is usually about OpenVMS and Tru64 Unix environments. Sorry about that.


Please be very careful when pulling any controllers from slot #2:
http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=PSD_OE040616_CW01

[Christian Bieletzki]

Uwe,

this is fantastic.
Where did you get this link from ...
Our controllers (before the change) had serial numbers that were affected by this.

Thank you very much for that information.
This might have been the root cause.

Best regards
Christian

[Uwe Zessin]

Hello Christian,
unfortunately, I don't remember where I got it from - I am glad I saved it away ;-)

[Christian Bieletzki]

The "inter-controller" link failure message seems to be related to ´heavy disk activity. If you do a manual failover and failback this works fine without error messages

Having a lage file copied across the FC connection the server does failover and failbacks.

Any ideas in this directio