We recently purchased two VNX7500s with FAST/Fast Cache, one for production and one for DR. Soon after purchase we realized a few things
1. SPs would spontaneously reboot.
2. There was an FCO stating that the SPs needed to be replaced (not because there was a physical problem with the hardware but because upgrading Flare codes in the future could be an issue) We were told that we received one of the first 50 or so VNX 7500s and they all required this FCO.
3. There was a bug in the Flare that caused reboots if Fast Cache was enabled on thin pools/LUNs
After having all the SPs replaced (as well as multiple components after the initial FCO was completed) and having upgraded to 5.31.000.5.011 (supposed to resolve Fast Cache reboot issue) we still seem to be having the issue of random SP reboots. We have yet to even re-enable FAST/Fast Cache on our production array until we are sure it is stable.
Is anyone else experiencing this issue?
There have been multiple discussions with EMC support and engineering. The first few go arounds it was an issue with FAST and Fast Cache being enabled. Even though the SPs had been replaced as part of an FCO we still ended up replacing more component parts, CPU, etc. Latest word is that only a handful of customers have experienced this issue and there are two primary causes: 1. code bug in RP splitter which causes single sp reboots during high load 2. faulty iSCSI driver (can still occur in systems not leveraging iSCSI).
.012 should fix both issues, due 9/5/2011
my VNX5700 was installed a couple of week ago and CE upgraded it to 5.31.000.5.012 ...by the way my 5700 just rebooted on its own yesterday and it was not even in use yet. This brings back very bad memories of CX600/700, i was hoping Clariion got better but so far not impressed.
On the other hand we've had a VNX7500 in production for well over a month now with significant load going on it (we are migrating an entire CX4-960 onto it) and have not had any such problems. Can't tell you the code level we are running off the top of my head and I'm not going to bother logging in right now, but it was whatever was current about two months ago. We haven't done a code upgrade on it since.
EMC released an upgrade for the RecoverPoint splitter on the VNX. We upgraded a couple of weeks ago. The arrays current config: Block OE is 05.31.007.5.011 and RP Splitter is 05.31.007.6.001. We haven't had a spontaneous SP reboot since the upgrade of the splitter code.