Hi,
Just wondering if anyone has experienced this. We have a couple of PeR720s connected to EQL41xx units via the above 6224 (stacked). The PeR720s are part of a Hyper-V cluster (Server 20112 R2).
For a while now, the cluster has been falling over- every 2/3 weeks. the solution up till now seems to have been to carefully shut down "unnecessary" guests, move the critical ones to one host then reboot each host in turn.
This didn't work the other day and cause chaos- nothing came back properly, the guests crashed etc. I noticed- for the first time- that the shared host networks were all over the place- hosts couldn't ping guests on iSCSI networks etc and vice versa, or pings were intermittent.
I rebooted the 6224 stack and everything was much more stable- zero packet loss, cluster behaving nicely, everything much faster.
Has anyone noticed 6224 units freezing up like this? They're running firmware 3.3.11.2 which was only released in July '14 so they can't have been up more than 7 months since the upgrade and a reboot.
This is obviously serious; the cluster is in production, and- if it is switch related- has to be fixed, we can't be rebooting the stack every couple of months (besides which, automatically rebooting this stack would be tricky).