[Beowulf] Anyone having IPMI problems on Intel S3200 series
Nifty Tom Mitchell
niftyompi at niftyegg.com
Fri Apr 17 15:31:17 PDT 2009
On Wed, Apr 15, 2009 at 10:09:15PM +0100, John Hearns wrote:
> 2009/4/15 Perry E. Metzger <perry at piermont.com>:
> >> Unfortunately, every once in a while, the IPMI BMCs on my test systems
> > simply stop talking to the network.
> I've seen this on Supermicro. As you say the cure is normally to do a
> cold reset from the OS on board,
> but as Greg says sometimes you have to do a complete power cycle.
> Best adice is to keep your firmware as up to date as possible.
Firmware updates have bugs....
As long as "possible" implies testing, soak time and review
this is good advice. So is "if it ain't broke don't fix it".
The most important time to check for versions and updates is after
a service call. Almost all the one off replacement mother boards,
raid and network cards I have seen have very old or very new firmware.
Since 'we' in the cluster community run Linux and most vendors have
'tools' that are limited to windows do not let the service rep escape
until he has tidied up the new hardware and perhaps built you a USB key/ CDROM
to update other boxes should an update be needed. Major and critical
updates are often quietly pushed into the spares inventory but customers
are not notified.
T o m M i t c h e l l
Found me a new hat, now what?
More information about the Beowulf