…but it says it does. It even reports that the mirroring completed successfully and that the volume status is “Synchronized” when the remote end in fact contains nothing but garbage data.
This is the result of what was described to me as a regression in a bad firmware release, but it’s unclear to me from my discussions with IBM exactly how far back this issue goes. I’m grateful that we didn’t find this in the middle of a production DR failover, but it’s completely ridiculous that an enterprise storage vendor allows such a serious data loss issue into a real release.
This is supposed to be fixed in a firmware update already GA’d but not on the website yet, but I’m awfully hesitant to actually use these large LUNs until IBM hashes out their support for them a little further. I’m not looking to be burned with the exact same thing with a different premium feature.
0 Responses
Stay in touch with the conversation, subscribe to the RSS feed for comments on this post.