I have performed block updates on a number of 5300's and a 5500 to 05.32.000.5.006 and 011 with no problems.
However, this will only be a problem if you are running 5.31.000.5.012 or older and is only an issue if you are running auto-tiering configurations.
Workaround if you are running 5.31.000.5.012 or later is to first upgrade to 5.31.000.5.726. Then stop data relocation on any existing pools that is running and then disable auto-tiering. Finally, you can upgrade to 05.32.00.011 with no issues.
Once the update is complete, you can re-enable auto-tiering and start relocation. Complete details can be found in the primus link below.
This is no longer a problem with Flare 32. The inital problem was with the release of Inyo code 5.32.000.5.006 when upgrading from a prior release with the RP splitter enabler installed.
This problem has been corrected in 5.32.000.5.008, so skip directly to this version and you will be fine including your existing RecoverPoint configuration.
The full support article can be found by looking up Primus # emc298607
Don't upgrade to 32 until after january 2013. There are several bugs that are being addressed and wont be available until SP1 comes out for it. Trust me. we found out the hard way.
We upgraded 2 arrays both 7500's. One of them has had 2 issues.
1. When trying to register wwn's or work within the initiators screen...info loading hangs for 5-10 minutes. Once loaded any changes will force java to eat cpu and Unisphere will be unrecoverable. Had to work w/ SME's at EMC to design an off-array client not yet accessable to public to fix the issue. Basically removing all the new code and reverting to old 31 code for the "host connnectivity" process. The fix is supposed to be out in bundle Jan 2013.
2. We have had SP's randomly reboot. bug check code E115801D. We are still waiting on engineering to see what that code means. It didn just bring down 1 SP... it brought down the other SP's connections for 9 drives. (dont ask me how)
Spindleguy
1 Rookie
•
14 Posts
2754
2
Posted October 17th, 2012 17:00
I have performed block updates on a number of 5300's and a 5500 to 05.32.000.5.006 and 011 with no problems.
However, this will only be a problem if you are running 5.31.000.5.012 or older and is only an issue if you are running auto-tiering configurations.
Workaround if you are running 5.31.000.5.012 or later is to first upgrade to 5.31.000.5.726. Then stop data relocation on any existing pools that is running and then disable auto-tiering. Finally, you can upgrade to 05.32.00.011 with no issues.
Once the update is complete, you can re-enable auto-tiering and start relocation. Complete details can be found in the primus link below.
Hope this helps.
http://knowledgebase.emc.com/emcice/documentDisplay.do;jsessionid=3FD4105CA058A1496BEF8480026E0691?docType=1006&clusterName=DefaultCluster&resultType=5002&groupId=1&page=&docProp=$solution_id&docPropValue=emc299619&passedTitle=null