This post is more than 5 years old
261 Posts
0
875
question on centera code upgrade
We have a Centera (EMC Centera Gen4) replication set in our production environment. We 're going for a code upgrade ..I 've two questions:
1) Now , we already have a disk failure in one of the nodes and an internal switch failure . My question is what is the risk when another disk failes ? I know , it depends on which disk has failed,but, if i 've to know what disk failure ( if second disk has to fail) can risk the primary cluster?
Basically , i 'm looking for what can jeopardize the Objects(data) in this scenario.
2) Also, What can be the backput plan set up when something goes bad.I understand , doing Code upgrade doesn't call for an outage. Just in case.
What should be my make out plan if the primary cluster goes down during code upgrade?
3) what are PEA files? what 's the use of it?
4)what is the consequence of replication being stopped, i see , repication from target to primary cluster is "on", however.
Data is stored with mirror protection
I will very much appreciate if some one can shed some light on this.
Thanks in anticipation!
holgerjakob_c0722c
337 Posts
0
March 1st, 2011 08:00
Hi
we manage 70 systems for different customers and are confronted with these questions quite often. My responses are based on my experience:
1) Upgrades are only supported on healthy clusters. You will need to engage support to repair your system first.
2) Upgrades can be performed without disruption. Downgrades are not possible. Make sure your applications are compatible with the new code prior to upgrading.
3) .pea Files are used by the applications to get access to Centera. No need to change them because of the upgrade.
4) Replication will be paused during the upgrade then resumed again. All writes during the upgrade will then be replicated.
Best regards, Holger