Hi Jeremy,
Thanks for this
We have problems in (I think) 3 of the 4. The first one that went (and
we replaced the disk) was the machine that we use to cache/serve the
data for the RTM. We thought "OK, fair enough, that machine does get a
pounding" so we replaced the disk ourselves. However over the last week
or so we have started to see errors on the machine that is our dcache
front end (oh, the joy that will cause if it falls over) and machine
that is our main UI. The only one of the four that I don't believe has
produced disk errors (so far at least) is the machine that runs our
production RB. This now gets a heavy load (thanks to the CMS JobRobot)
but was only lightly loaded for quite some time.
At least we are getting good use out these machines ...
All the best,
david
Coles, J (Jeremy) wrote:
> Hi David
>
> See: http://www.gridpp.ac.uk/tb-support/warranty.txt - 36 months
> warranty on all components. As the machines were handed over to the
> institutes it is up to you to contact Streamline to report the problem
> and get replacement. The contract suggests a 2 working day turnaround
> but I would be interested to know how you got on. Are all four machines
> now showing problems and did they all start at about the same time?
> Perhaps it relates to what you feed them or how they get used!
>
> I'm aware that one other site did have initial problems but I do not
> recall which site - I'm sure the issue was caught in the TB-SUPPORT
> archives but will not check just now.
>
> Jeremy
>
>> -----Original Message-----
>> From: Testbed Support for GridPP member institutes [mailto:TB-
>> [log in to unmask]] On Behalf Of Kant, D (Dave)
>> Sent: 20 October 2006 13:31
>> To: [log in to unmask]
>> Subject: Re: GridPP provided machines
>>
>> Hi David,
>>
>> I'm using two of these machines at RAL for accounting.
>> We run MySQL with RGMA and have about 15million job records.
>> The load on one of these machine is mostly high > 50%
>> I've not had any issues with the disks.
>>
>> Dave
>>
>> =========================================================
>> Dr Dave Kant
>> CCLRC eScience Department Phone: (+44)|(0) 1235
>> 778178
>> Rutherford Appleton Laboratory Fax: (+44)|(0) 1235
>> 446626
>> Chilton, Didcot, Oxon, OX11 0QX, UK Email: [log in to unmask]
>> ==========================================================
>>
>>
>> -----Original Message-----
>> From: Testbed Support for GridPP member institutes
>> [mailto:[log in to unmask]]On Behalf Of David Colling
>> Sent: 20 October 2006 13:13
>> To: [log in to unmask]
>> Subject: GridPP provided machines
>>
>>
>> Hi,
>>
>> We are starting to have a lot of disk problems on the four machines
> that
>> were given to us by GridPP. So I would be interested to know:
>>
>> 1. If other people are seeing (have seen) similar problems or whether
>> ours are unique ... perhaps caused by an unusually pulse of neutrons
>> from the nuclear fusion device at the other end of the corridor.
>>
>> 2. If anybody knows what the warranty status is. I should know as I
> was
>> involved in the procurement, however I disposed of my copies of the
>> documentation a few weeks ago when I re-arranged my office. I seem to
>> remember that it was "3 years return to base" that Streamline offered
> us.
>> All the best,
>> david
|