4 weeks ago
We just recently purchased some new AFF8080's and are having some difficulties with ADP on 8.3. The two HA pairs that we purhcased had come with the factory pre-configured cluster. This configuration didn't seem ideal to remove them from the current factory cluster and then add them to our production cluster. We were concerned about trace remnants of the previous cluster showing up in our new cluster. Well we went through and performed a wipeconfig on the nodes and everything seemed good until we found some unknown disks. From troubleshooting and working with support we found that the disks hadn't been properly cleaned up due to possibly the partitions not being cleaned up during the wipeconfig. We found some documentation after the fact that showed how to go through and properly re-initialize on 8.3 with ADP.
We were able to go and cleanup the unknown disks, and everything actually looks good now except for one little thing. The root partition on one of our nodes is spanned across multiple disk shelves. It doesn't show any signs of issues, but I'm concerned whether this could cause issues in the future? We're wondering whether we should take the time again now to go through and re-intialize that specific HA pair or if we're potentially just over analyzing this?
Any advice is appreciated.