HA unavailable

[tino]

Hello,

High availability is unavailable in my grid due to the following error.
---------------------------

HA check failed: 1 catalog class with shared volumes has degraded volumes [/system:NET.usr]
----------------------------

I tried to repair this volume and getting the following error.
------------------------------

failed to start repair of volume _cat.system.NET.usr - return code 12032

-----------------------------

Please help me to resolve this issue

[PeterNic]

Hello tino,

What version of AppLogic did this occur on?

Also, do you see anything in the controller log ("log list"); if you have maintainer access, you can see more details in /var/log/messagses on the controller, as well as on the server on which the new stream is likely to be created.

A suggestion: try disabling the server on which the new stream is likely to be created - if the repair proceeds and finishes normally, it is likely you have a "stuck volume mount" on that server (please post the output of "vol info -b" for that volume if you need help in determining where the new volume mirror will go). Rebooting the server will likely repair the problem. Our support may be able to help fix this without a reboot.

Best regards,
- Peter

[manojsamtani]

Hi Peter,

We are facing the same issue log file says:

failed to start repair of volume _cat.system.PS8.usr - return code 12032


Please provide help on this......


--
Manoj Kumar

[PeterNic]

Hi Manoj,

What version of AppLogic do you use, with what hotfixes? Have you tried any of the suggestions above?

Also, if you have maintainer access, can you try the following from the controller (as a maintainer) and let me know what the output was:
3tctl repair sv _cat.system.PS8.usr
(this will try to force repair the particular volume and give you any immediate error messages; also check the system log after it, as well as the system logs of the two servers where the streams are -- those are shown with "vol info /system:PS8.usr" executed as a regular user.)

I am working on the assumption that you - and tino before you - have hit the "stuck mount" problem in which a volume cannot be mounted, and therefore cannot be repaired. If this is the case, you might experience other problems - e.g., starting apps that involve the server on which the volume mirrors reside. The problem is easy to fix and our support knows how to do it - please open a ticket and immediately request escalation to L2 support.

If you or tino resolve the issue, it will be very kind to post what the problem/solution was.

Best regards,
- Peter