[Wg-test-framework] oseleta* (Dell R415) problem, offered free upgrade/sidegrade to R630

[Ian Jackson <ian.jackson@xxxxxxxxxxxxx>]

The test lab has four Dell R415s, in two pairs with different
configurations.  The nocera pair (nocera0, nocera1) basically work
fine and have been in service for some time.

The oseleta pair have some problem which causes them to hang during
the BIOS POST about one time in 20.

Dell have replaced several components; they have tried to repro the
problem in their lab, under my instructions.  Credativ and I have
narrowed the buggy configuration down to just the server (with no
network or keyboard connected) and Credativ reproduced the trouble on
Yogesh's workbench, so we can rule out problems with the datacentre
power (or serial provision).  However, Dell haven't been able to repro
the problem.

Dell have now decided to cut their losses.  They have offered to swap
the two problematic R415s for two R630s.  The technician writes:

  I'd like to offer exchanging both systems at this point since we
  have not been able to replicate the issue here in our lab.  Since
  the R415s is no longer built, I would like to offer R630s in their
  place to compensate for an AMD to Intel change as we do not have any
  new AMD systems.  If you would like to see the design of what the
  new system will include, I can create a build for the systems and
  send it to you to look over.

I am inclined to say that we should accept this offer.

The switch from a pair of AMD x86 boxes to a pair of Intel ones is not
ideal but Lars tells me we expect to be getting a second rack, and we
can rebalance as desired when we do the procurement for that rack.

Opinions ?  In the absence of objections I intend to ask Dell to
please, yes, send us a spec to review.

Thanks,
Ian.

[1] If I have got my records correct, the troublesome R415s (oseleta*)
have this configuration:
  
http://xenbits.xen.org/gitweb/?p=people/iwj/colo-for-testing.git;a=blob;f=avail/Item+3A-Quotation%23112514-1+DELL+1U+DUAL+AMD+4133.doc;h=a6878dbacf3cb38b1406b1af129790c60e20ed1f;hb=HEAD

_______________________________________________
Wg-test-framework mailing list
Wg-test-framework@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/cgi-bin/mailman/listinfo/wg-test-framework