Storage Boards Cleanup
To make it easier to find information about HPE Storage products and solutions, we are doing spring cleaning. This includes consolidation of some older boards, and a simpler structure that more accurately reflects how people use HPE Storage.
Disk Arrays
cancel
Showing results for 
Search instead for 
Did you mean: 

Help. All Disk Repliction group failed. Few space disk available.

laurent49
Occasional Advisor

Help. All Disk Repliction group failed. Few space disk available.

Hi,

We have troubles for 15 days with our 2
eva 8000 ( 35 To x 2 ) ... Today we have an other big problem :

All disk group replication are now failed.
They are logging, not merging.

We have troubles since I tried to increase a vdisk from 1To to 1To5 for radio images ( hospital ) ... after 15 days it's still in progress. Why ? We undertood but too late that
we have not enough space disk ... only 300Go available, and this type of command is strong for EVA.
Today I have less and less disk space availaible on my Disk group FATA 500 and Fiber 300 ...
We tried do delete DiskGroup réplication, but we have : storage modification exeption : dr replication failed ( while removing storage volume ) : Cannot modify object while DR group is logging.
If only we could delete vdisk replicated we could get some disk space : but how to do that ?

When I click on disk group I have : logging not merging.

Has anyone a solution for this case ?
Delete vdisk replicated on dr group which is logging ( not merging ) ?

I am affraid about this status, because it's
critical for all vdisk ...

I am not expert on eva 8000 ...

my e-mail : laurent.patrick@laposte.net

Thanks for any support.

Patrick.
5 REPLIES
Víctor Cespón
Honored Contributor

Re: Help. All Disk Repliction group failed. Few space disk available.

The disk group has "Disk drive failure protection" set on Single. This is taking 558 GB. If you set it to None, you'll recover those 558 GB, which will come very handy on that disk group which is 99.5% full (a very bad idea, by the way).

The DR group is logging, is the data replication link up?

This is the advisory for expanding a vdisk in a CA environment:

http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=c01456501

See how it behaves with more space available...
laurent49
Occasional Advisor

Re: Help. All Disk Repliction group failed. Few space disk available.

I prefer to keep single on disk protection.
We have to many vdisk on production. It's too critical.

When I click on the vdisk I am trying to increase, I have :
Capacity
requested : unknown GB
allocated : unknown GB

The document http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=c01456501
is interresting, but I have unknwown GB, so should I follow instruction on this document ?

You had a good idea regarding replication link : has DR group is logging is the data replication link up?
Could you tell me how to check that on my two
EVA 8000 ? How to check that data replication link is up ?

I view different errors on event viewer :
A Data Replication Log has been reset due to insufficient Disk Group capacity; The Data Replication Destination has been marked for a Full Copy.
Corrective action code: 62

Is it interresting to launch a full copy on a DR group which is logging ( not merging ) ?

I found this error :
Management of a storage system has been suspended due to an inability to communicate with the array controllers. Management of the storage system will recover automatically when the system is able to respond.

I attached a word doc with errors.

When I click on a Disk Group ( data replication ) , I have :
Operationnal state : Failed
DR role: Destination
Failsafe mode: Disabled
Replication path: Controller: Offline
Port: Not available

Patrick.





laurent49
Occasional Advisor

Re: Help. All Disk Repliction group failed. Few space disk available.

On RSM, I found this :
Warning :Connection Failed, resumed, logging.

See print screen on attach doc.

McCready
Valued Contributor

Re: Help. All Disk Repliction group failed. Few space disk available.

At this point, in my opinion, you may need some downtime to un-confuse your EVA's, as it appears you got yourself into some black hole that you will not get out of until enough space is available. I would highly recommend getting your HP support involved to see what the best method going forward would be.

If that is not possible, with my back up against the wall, I would,
- do a "Capture configuration" in case you mess up something with the below
- stop all replication and delete the DR groups; once they are "logging", they will be using up the rest of your free space for their logs until the log limits have been reached, and the full-copy flag is set. Do not do this if you do not think you can recreate your replication setup.
- delete any vdisks not needed on the array with the space problem to free up space, and wait until any leveling is completed to get the total amount of free space you'll have. I would recommend keeping at least 5% of your array available, that amount varies inversely with the size of your array.
- restart your replication to get that back up, and wait until you are no doing full copies. You may need to delete the old replicated-to vdisks on the remote side.
- determine if you really have enough space available to expand your disk, leaving room for your DR group logs in addition to the 5% or so space. You may have to stop/start replication to do this.
- notify management of what you plan to do, and what the impact might be (i.e. no replication for a few hours/days, etc.)and be sure you understand the implications.

But take the above with a several pounds of salt - something may be wrong with the EVA that requires professional HP assistance
check out evamgt.wetpaint.com and evamgt google group
laurent49
Occasional Advisor

Re: Help. All Disk Repliction group failed. Few space disk available.

Hi,

Today we had HP support.
They connected on our site through webex session.
They changed our zonning to stop replication between bays and successfully removed a DR group replication which was increasing ( dr group which has a bad vdisk ).
But we still have a vdisk with a status of unkwnown size. HP will probably come on our site and restart controllers ...