Unsolved
This post is more than 5 years old
615 Posts
0
4234
Need help getting one of my PS4100 members back online.
I have an issue with a ps4100 at firmware 9.0.3.
It is a member of a group.
There are two identical members in the group.
They are both usually in the same production pool.
I needed to physically move one of the members so I moved it to a maintenance pool.
Once completely in the maintenance pool and no longer having any active traffic, I shut down the member and physically moved it.
Upon powering it back up it seemed like it took longer than normal for the cm0 controller to be recognized and rather then being patient I pulled the controller and reseated it. In doing so I believe I corrupted the data on its SmartCard. I did not know this at the time.
In fiddling with the member trying to fix it I now have this member in a state where if I try to restart from cm1 I get the following error:
Rebooting the active controller
reboot: A RAIDset is initializing, dirty, or faulted beyond recovery.
The array cannot be rebooted or halted in this state.
I have console connectivity to both controllers.
What is the best route to get this member back into a healthy state so I can bring it back into the production pool?
Do I need to reset it and rebuild it?
Anonymous
5 Practitioner
5 Practitioner
•
274.2K Posts
0
November 1st, 2017 12:00
Hello,
What you should have done is after moving it to the maintenance pool select "Delete Member" This would have removed it from the Group DB and reset the RAIDset on that member.
The error is indicating there is a problem with the RAIDset. It has too many drives failed to bring the RAIDset up. If the controllers boot up, then the compact flash card is fine.
I would suggest powering off the array, then reset all the drives.
On the active controller, please run: GrpName>raidtool
Regards,
Don
Anonymous
5 Practitioner
5 Practitioner
•
274.2K Posts
0
November 1st, 2017 13:00
Hello,
Either CM can be active, so if CM1 boots it should be the active CM. So raidtool should still work. The CLI>prompt means it can't read the configuration info from the RAIDset.
Re: Offline. No I can't do that. You will need to call in to get that response. You'll have to run it again once you get on phone with them.
For a fee you can get a one time support call now. They will help you triage it.
Regards,
Don
cameronredux
615 Posts
0
November 1st, 2017 13:00
Thanks Don.
All drives were healthy before the physical move and were fine when the member was running on cm1 before I fiddled with the member further trying to fix the cm0 issue.
I am at a state where I am stuck in CLI> on cm1 and the member will not boot from cm0 so I am unable to run raidtool.
The member is in an offline state now so I am not able to delete it from the GUI.
I am not able to delete it from the cli (when logged into the functional member) as it asks for a response from support but it is out of coverage.
At this point I would like to just delete it and reset it.
Would you be able to provide a response to this challenge?
Firmware version: V9.0.3
Challenge: 8934
Thanks!
cameronredux
615 Posts
0
November 1st, 2017 15:00
Thanks Don!
Both controllers were at CLI> so I was stuck :)
I was able to get a response code from support and deleted the offline member from the group and was then able to re-add it after resetting it.
It's initializing now in the maintenance pool and I will be able to bring it back into the production pool in a few hours.
Anonymous
5 Practitioner
5 Practitioner
•
274.2K Posts
0
November 1st, 2017 16:00
Hello,
Since the RAIDset is down, the CLI> prompt is completely normal. raidtool is an internal command.
Glad you got it back.
Regards,
Don