Same here. On our test ESXi5 host, we were seeing the same issues outlined in the KB article linked above, and on our NetApp controllers we saw frequent FCP Logins from the initiators that were associated with this host with only moderate I/O (I'd say anything >1000 IOPS, such as powering on a single VM, or deploying one VM from template). The posted workaround did resolve our issues, but I am wondering about the side effects of making this change.
↧