Unsolved
This post is more than 5 years old
39 Posts
0
3466
Issues in Mc_Retired
I have 2 10 node grids with the second setup at our DR site to recieve replications. Each grid is version 6.1.0-402. My question relates to the MC_Retired collection. On the primary node, in the MC_Retired collection, I have servers listed in there multiple times that aren't retired. I try to go in and clean them out by deleting them and get the following error:
Can you tell me what this error is and why its happening? Maybe you could also tell me why my clients are showing up in the MC_Retired collection when they aren't retired.
Second issue is in the same MC_Retired collection but on the replica grid. There are servers in this collection that have been gone for a long time and should not be in there. Plus I have multiple copies of servers in there too. My retention time is 42 days which I retain when I retire a client and I know darn well that this time has expired so in my opinion, they should be gone. Why is this and is it safe to delete the ones that I know should not be there?
Any help would be appreciated.
Thank you.
Sheila
Pratique2511
82 Posts
0
May 15th, 2013 04:00
Since, when are you getting the error ? This generally pops up when the grid is in Read-Only mode or replication job is running. Please check if any of the above is the reason.
You can retire the server when Grid is in Full Access and no replication job is running.
sheilaa2
39 Posts
0
May 15th, 2013 05:00
Thank you for the reply. I just started noticing this on Monday and get the error at multiple times during the day. Both grids are Full Access and there are no replication jobs running. Any other suggestions? My only other thought is that maybe it's time to reboot both grids to see if that clears up the issue.
Pratique2511
82 Posts
0
May 15th, 2013 05:00
May be try checking the unacknowledged events if there are. That might help.
sheilaa2
39 Posts
1
May 15th, 2013 06:00
Thank you so much. I will be opening a ticket with support so we can clean this up.
ionthegeek
2K Posts
1
May 15th, 2013 06:00
Each time a system is retired and re-activated, a new client entry will be created in MC_RETIRED with the date the client was retired as the suffix. Even if all the backups associated with the client in MC_RETIRED have expired, the client entry will remain unless it's specifically deleted.
If these client entries are not causing operational problems, there's no specific need to delete them. From the message you posted, it appears that there may be a client ID (CID) mismatch for one or more of these clients so if you wish to delete them, please open a ticket with support to get that straightened out. The team best suited to handle your issue will be the Capacity and Replication team since they deal with CID mismatch issues on a regular basis.
I would recommend against rebooting the system -- this will not make any difference to the problem at hand and there's always the risk of triggering a rollback if the shutdown procedure isn't followed closely.
J_H_
498 Posts
3
May 15th, 2013 12:00
what does the error say when you hit "view details" that will give you a bigger clue as to why you got the error.
and as ianderson says - I have noticed the same thing.
and I like it clean...
so the first of every month I go into backup management and look at each of the retired servers backup by date ranges (choosing a date back before I know we had Avamar).
If it still has valid backup images they will show up.
If they no longer have any valid backup images I delete it from MC_Retired (via the administration window)
I will then go to the replication grid and remove it from the replicated/mc_retired as well.
Less confusing to anybody else who might have to look at this when I am on vacation.
sheilaa2
39 Posts
0
May 15th, 2013 12:00
Exatcly! I like to keep this clean too. Why leave them hanging around? EMC is looking into the issue for me. Basically, I can tell you that they are all VM servers that are giving the error. I can also tell you that we've replaced our vcenter server which is why there are duplicates but I should be able to delete them whether or not there is backup data associated with them or not! When I click on the view details.... nothing but useless info:
sheilaa2
39 Posts
0
May 29th, 2013 07:00
Had to open a ticket. The tech went in and cleaned out all of the "orphans". What a mess! Then I had to manually remove them from the replica grid. Would be nice if this could be automated someway.