I hope someone will be able to enlighten me on what is happening : I have two identical GPUs, one is failing to validate all WUs, and the other is perfectly fine.
I've tested the failing card with other DC projects, MemtestG80 and MemtestCL, and I didn't find any error ...
So here is the list of the tasks with validate error:
I hope someone will be able to enlighten me on what is happening : I have two identical GPUs, one is failing to validate all WUs, and the other is perfectly fine.
I've tested the failing card with other DC projects, MemtestG80 and MemtestCL, and I didn't find any error ...
So here is the list of the tasks with validate error:
Ive noticed a lot more of these validate errors in the last 3 months & like Joe stated earlier i myself am no expert just curious. They dont seem to be cpu or gpu specific or am i wrong ?
Ive noticed a lot more of these validate errors in the last 3 months & like Joe stated earlier i myself am no expert just curious. They dont seem to be cpu or gpu specific or am i wrong ?
There's quite a lot of detailed information on error rates for various platforms for the various science apps in the opening posts of this thread. These errors have been around for quite some time (much longer than 3 months) and are being investigated. They seem to be OS specific - Windows hosts have much lower error rates.
Two
)
Two today:
262704444
262862245
NG
NG
RE: Two
)
Both of them have the same error message
Cheers,
Gary.
These are 12 days
)
These are 12 days worth:
http://einsteinathome.org/workunit/112033008
http://einsteinathome.org/workunit/111956773
http://einsteinathome.org/workunit/111860765
http://einsteinathome.org/workunit/111841571
http://einsteinathome.org/workunit/111837283
http://einsteinathome.org/workunit/111769599
http://einsteinathome.org/workunit/111741911
http://einsteinathome.org/workunit/111290389
http://einsteinathome.org/workunit/111689506
http://einsteinathome.org/workunit/111688769
http://einsteinathome.org/workunit/111537366
http://einsteinathome.org/workunit/111522011
http://einsteinathome.org/workunit/111467739
http://einsteinathome.org/workunit/111464474
http://einsteinathome.org/workunit/111407829
http://einsteinathome.org/workunit/111392309
http://einsteinathome.org/workunit/111368781
http://einsteinathome.org/workunit/111348576
http://einsteinathome.org/workunit/111269845
Joe
I hope someone will be able
)
I hope someone will be able to enlighten me on what is happening : I have two identical GPUs, one is failing to validate all WUs, and the other is perfectly fine.
I've tested the failing card with other DC projects, MemtestG80 and MemtestCL, and I didn't find any error ...
So here is the list of the tasks with validate error:
266671780
266684326
266690957
266704182
266707576
266723178
266728665
266732827
266744893
266747992
266837405
RE: I hope someone will be
)
I took a quick look and by my count that computer has 11 Validate Errors and 13 Validated results. Are both GPU's in the same system?
I am not an expert here just curious.
Joe
Ive noticed a lot more of
)
Ive noticed a lot more of these validate errors in the last 3 months & like Joe stated earlier i myself am no expert just curious. They dont seem to be cpu or gpu specific or am i wrong ?
X
RE: I took a quick look and
)
Indeed, valid results are from GPU 1 and CPU, and invalid result are all from GPU 0.
RE: Ive noticed a lot more
)
There's quite a lot of detailed information on error rates for various platforms for the various science apps in the opening posts of this thread. These errors have been around for quite some time (much longer than 3 months) and are being investigated. They seem to be OS specific - Windows hosts have much lower error rates.
Cheers,
Gary.
RE: Indeed, valid results
)
Which fairly strongly implies that your GPU 0 has recently developed some sort of hardware problem and needs to be checked out.
Cheers,
Gary.
Unfortunately, it doesn't
)
Unfortunately, it doesn't fail in any other application I tried (MemtestCL, MemtestG80, SETI, Folding).
And there is no output in E@H stderr that could help to find out what is wrong :(