P4500G2 error while adding in cluster Centralized management console

[Waseem_2]

Recently, one of my customers experienced a hardware fault in an HP 4500 G2 Storage system. The hardware issue has since been resolved. However, after the fix, when attempting to run the 'Repair Node' option in the Centralized Management Console, a warning appears stating that the cluster configuration will change from a Multi-site cluster to a Standard cluster.

A RIP node has appeared in the cluster group (the cluster consists of 4 nodes, and one of them—the RIP node—had a hardware fault). Although the hardware issue has been resolved, the node is not showing as part of the cluster. When we attempt to add it back to the cluster or use the 'Exchange Node' option, we receive a warning stating that the cluster configuration will change from a Multi-site cluster to a Standard cluster. Please help

[Assen]

Hi Waseem_2,

Thanks for the detailed description. It sounds like the hardware issue has been resolved, but the cluster is still seeing the affected node as a RIP (Removed In Progress) node, which is why you're encountering that warning when attempting to repair or reintroduce it.

Before proceeding further—just to double-check—did you follow the recommended HPE procedure for repairing a node in a Multi-site StoreVirtual cluster? This helps ensure the node is correctly re-integrated without triggering a change from a Multi-site to a Standard cluster. Here's the link to the official guide:

HPE Support Document – Repairing a Node in StoreVirtual Multi-site Cluster

Also, it’s worth noting a key difference between the Add Node and Exchange Node options:

Add Node: Adds the system as a new node to the cluster. If the old node is still marked as a RIP node, this can change the cluster’s configuration type.

Exchange Node: Replaces the existing placeholder (RIP) entry in the cluster with the repaired node, preserving the cluster configuration and avoiding unexpected changes.

If you still see the warning when using Exchange Node, it might suggest the node isn’t being fully recognized as the original or expected member. Following the correct repair procedure should resolve this.

Let me know how it goes.
Maybe post also some screenshots of the CMC to get a better overal picture of the situation.

Best regards,
Assen

[Waseem_2]

Hi Assen,

I shut down the node and replaced the cache module. After the replacement, I encountered this issue. Please find the attached picture for reference. Thank for share the guide.. I will check and let you know the outcome.

[Assen]

Hi Waseem_2,

thank you for the screenshot. It looks like you have repaired the system and have added it back to the Management group (step 5).

Here are the steps to repair a broken storage system, be it cache or brocken raid.

https://support.hpe.com/hpesc/public/docDisplay?docLocale=en_US&docId=emr_na-kc0112618en_us
1. stop store from cliq only needed if San/iQ version is before 10.5
2. stop manager from CMC / make shore you have enough managers to keep quorum!
3. put into repair CMC / this is a storage issue
4. reconfigure raid under raid setup vsa, while in available systems
5. add back to existing MG
6. start manager again on VSA
7. add to correct site if applicable
8. edit Cluster > Exchange System (important point is to Exchange and not to add or swap!) / exchange IP placeholder with repaired node
9. wait for restripe
10. when volumes normal again, all done

In your case as you have already added the storage to the existing management group (step 5) and If you think that all is ok now and node can be added back follow the steps from step 6 in your case.

Start a manager on the KMNode4 if there was a manager running berfore the repair and addit after that to the correct site if applicable.

Than just go through the step 8.

That should do it.

Cheers,

Assen

[Waseem_2]

Hi Assen,

I have a customer session today. Since I only replaced the cache module in the system, do I still need to run the 'reconfigure RAID' operation? please advise.

Till now I have performed below steps. 

• Power down the system.
• Replace the cache module and cache battery.
• Repaired the storage system.

[Assen]

I would say you only need to reconfigure raid if the raid is broken or you get some other error regarding raid or cache.

Cheers,

Assen

[Waseem_2]

Hi @Assen, I need urgent support

Currently in a session with customer and trying to perform the steps which you suggested earlier. Node is showing unassigned to site and getting an error "cluster configuration will change from multsite to standard custer while trying to assign the site to the node.. can you please advise.

and also no option is available to add the node in existing management group.. please advise.

[Assen]

Hi Waseem_2,

thank you for the screenshots.

You should carefully read and follow the steps that I sent. From the screenshots you sent, the node is already in the managment group, hence no option to add it! Now you have to follow on from step 7.

You have to add the node to the correct side first and than edit cluster and select exchange RIP ip with repaired node.

1. stop store from cliq only needed if San/iQ version is before 10.5
2. stop manager from CMC / make shore you have enough managers to keep quorum!
3. put into repair CMC / this is a storage issue
4. reconfigure raid under raid setup vsa, while in available systems
5. add back to existing MG
6. start manager again on VSA
7. add to correct site if applicable < You are here.In the site config select "add" and add the node to the correct site. I think you can ignore the warning that pops up. From what I see you have added the node to correct site.
8. edit Cluster > Exchange System (important point is to Exchange and not to add or swap!) / exchange IP placeholder with repaired node
9. wait for restripe
10. when volumes normal again, all done

If all still fails you can always open a case with HPE support.

Best,

Assen

[Waseem_2]

Hi Assen,

I have performed the below steps:

1. stop store from cliq only needed if San/iQ version is before 10.5
2. stop manager from CMC / make shore you have enough managers to keep quorum!
3. put into repair CMC / this is a storage issue
4. reconfigure raid under raid setup vsa, while in available systems (this step has been skipeed bcz data backup not available)
5. add back to existing MG
6. start manager again on VSA
7. add to correct site if applicable < You are here.In the site config select "add" and add the node to the correct site. I think you can ignore the warning that pops up. From what I see you have added the node to correct site.
8. edit Cluster > Exchange System (important point is to Exchange and not to add or swap!) / exchange IP placeholder with repaired node

After performing the above steps up to step 6, when I proceed to step 8, I receive a warning stating, "Cluster configuration will change from multi-site to standard cluster." Could you please advise on this? A screenshot is attached for reference.

The same warning I getting when I am trying add this node to site (Step7)

Unable to log get HP support because this product is out of support.

[Assen]

Hello Waseem_2,

as I said, please add the node in step 7 to correct site, as you allready showd in this screenshot. Klick on OK and ignore the warning.

Then do step 8, exchange the repaired system into the cluster, as you showed in the last screenshot and klick ok. If there is a warning about site change just ignore it.

You can still open a support case even if the product is out of support; however, you will need to request a quote and accept it before support can begin working on the issue.

While this remains a solid product, I strongly recommend that your customer consider a replacement. Options like an on-premises Alletra system or a GreenLake consumption-based solution may be worth exploring.

Best regards,
Assen

[Waseem_2]

Hi Assen,

Thank you so much.

Let me discuss this with the customer. Hopefully, it will not have any impact on the cluster configuration.