GSP PDC Clarification.

[joe_91]

Team:

Here are my questions...
1. What are these GSP PDC?

2.What would be the effect of having older levels of these? I have 41.36 PDC

3. If the motherboard is replaced, then would the PDC and GSP Version change?

4. what is the process to upgrade the GSP and PDC?

Please Help.

Thanks

Joe.

[Stf]

Hello,

About GSP :

http://web.tampabay.rr.com/batcave/GSPinfo.htm

Stf ;-)

[Robert-Jan Goossens]

Joe,

GSP = Guardian Service Processor

What kind of server do you have (l-class or N-class) ?

Regards,
Robert-Jan

[harry d brown jr]

1. What are these GSP PDC?
GSP manual: http://docs.hp.com/hpux/pdf/A5201-90017.pdf

2.What would be the effect of having older levels of these? I have 41.36 PDC

The newer PDC's fix things that are broken.

3. If the motherboard is replaced, then would the PDC and GSP Version change?

Potentially.

4. what is the process to upgrade the GSP and PDC?

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=566357

live free or die
harry d brown jr

[Bharat Katkar]

hi
find attached doc on GSP.
regards,

[joe_91]

Thanks for the replies. I have N class.

Thanks

Joe

[Robert-Jan Goossens]

Hi,

You can install an OS patch to upgrade your GSP. Below you can find the link for the 11.11 patch, there is also a 11.0 patch.

Patch Name: PHSS_30724

Patch Description: s700_800 11.11 N4000/rp7400 43.43 PDC Firmware Patch

http://tinylink.com/?w26mAVToF4

Regards,
Robert-Jan

[Richard Darling]

Hi Joe,
I kept getting Event Monitor notification information:
Event # : 32
Severity : SERIOUS
Service processor not responding

After installing the patch to upgrade the GSP the error stopped, and apparently it was just a false alarm because the GSP was outdated.

I get the feeling from your message that you are reluctant to upgrade unless necesary.

You didn't say why you were asking about this, but if you are getting false alarms there is a workaround the upgrade.

The following is how to prevent EMS from reporting an event (event 32 used as an exmaple):

Edit the file /var/stm/config/tools/monitor/default_dm_core_hw.clcfg
and look for these lines:

# EVENT_PE_SERVICE_PROC_DEAD:
EQ:32 :SERIOUS :TRUE :1440 :60 :4 :NONE :NO_OP:NO_OP:NONE

Apply one of the workarounds.



Change the severity of the event to prevent messages from being logged
to syslog.log and e-mail being sent to the root user:
EQ:32 :INFORMATION :TRUE :1440 :60 :4 :NONE :NO_OP:NO_OP:NONE


Prevent the event from being reported altogether:
EQ:32 :SERIOUS :FALSE :1440 :60 :4 :NONE :NO_OP:NO_OP:NONE


Change the threshold:
EQ:32 :SERIOUS :TRUE :1440 :3000 :2 :NONE :NO_OP:NO_OP:NONE


Instead of reporting the event if the problem occurs 4 times in 1 hour,
we only report it if the problem occurs twice in a little over 2 days
(50 hours).