node_therm_state hw_node:0 Node 0: SBB Canister 0 at WARNING

[glanzmann]

Hello,
we have the above error message followed by:

node_therm_state hw_node:0 Node 0: SBB Canister 0 at WARNING level (50 C)
node_therm_state hw_node:0 Node 0: auto-fix event

Has someone seen this event and can elaborate what is happening here? I assume it is the drive
cage of node0 of 7400, is that correct?

Cheers,
Thomas

[Cleanur]

Yes.
Assuming you don't have a cooling problem in the DC then from memory the alert is typically a bug introduced with the newer thermal detection and shutdown code, basically the alert triggers too early. A patch was in the works so it may already be available if you log a call..

[hdtvguy]

This is a known bug in 3.1.3 and 3.1.3 MU1. I have been fighting with support as we were told the bug was fixed in MU1 and we had our 7 array upgraded only to find the bug will not be fixed until 3.2.1. We have no intention of going to 3.2.1 for some time so we are fighting for a patch.

[glanzmann]

Hello guys,
thanks for the all-clear. We checked the machine room and the temperature is okay. So it seems
that triggered the same bug. We're running 3.1.3 MU1.

Cheers,
Thomas

[Leif]

I can see a hint about a soloution of the false temperature alarms in patch 09.

HP 3PAR OS 3.1.3 MU1 P09 includes:
• Issue ID: 118279, 114735
• Description: An incorrect alert is resolved by changing the alert to post at the correct
node/canister temperature warning level.