ProLiant Servers (ML,DL,SL)
cancel
Showing results for 
Search instead for 
Did you mean: 

BL20p G3 + ilo 1.70 loss of service

Tom Pepper
Occasional Contributor

BL20p G3 + ilo 1.70 loss of service

All:

Have six BL20P G3s in a single rack, wired correctly. Frequently (once a week) one of them (random one) fails to respond to iLo requests -- indeed, any network request. Failure begins as loss of HTTP/SSH services, and eventually culminates in loss of ICMP. Only way to restore is to physically pull the blade out and replace. Yuck.

There doesn't seem to be a reason why it happens, it happens regardless of whether or not any activity is happening at all to the iLo. I've seen it happen on completely idle MPs.

ILO is cabled directly into GbE interconnect on management VLAN. other MPs are responding fine.

It would be *really* nice if there was a method to send an iLo restart through the enclosure or from other MPs in the same bladesystem, especially for blades not running a sanctioned OS and lacking insight management drivers. This particular instance happened while installing vmware ESX 2.5 (which seems to hang during the final config phase) so now I have a dead blade I have to pay someone $100 just to go pull out and replace.

help,
-Tom