Getting a bunch of 'validate errors'

http://einstein.phys.uwm.edu/results.php?hostid=790150

Project was reset but errors continue.

Rosetta & Seti on same box show no problems. Suggestions?

Thx.

You've got infinite numbers in the result files. This usually points to a hardware problem, CPU getting too hot, memory fault etc.

BM
____________
BM

Ok thanks, i'll chalk it up then to a project which stresses hardware more than the others mentioned and/or has more stringent verification.

Correct, this project is probably among the strictest in verification. Bernd is one of the developers, and knows his software very well, so he's given you the exact reason you are getting validate errors. The issues could result from any of a number of things.

I would do the simple things first. Clean, clean, clean. Make sure every fan is clean, the heat sink is clean, the heat sink is not loose, etc. Then do some of the simple tests, like Memtest86+ and Prime95.


____________

Ok thanks, i'll chalk it up then to a project which stresses hardware more than the others mentioned and/or has more stringent verification.

I also get 'validate error:

http://einstein.phys.uwm.edu/results.php?userid=64307

What's wrong?
____________

You've got infinite numbers in the result files. This usually points to a hardware problem, CPU getting too hot, memory fault etc.

I am getting validate errors also from my new woodcrest(stock no ock) and receiving wu's that have had validate errors. Now I haven't had but 1 validate error in almost 2 years here What Gives? Should I abort the ones that have not validated? I hate crunching long units when I see a problem ahead of time....I don't think it is the hosts ,I think Bernd you need to look at your validator as too many of these are popping up at once.If this continues I will suspend the project and crunch elsewheres. Please respond... Thanks JR
____________

I do get Validate errors as well.
Strange that 4 machines have validate errors on the very same WU and now two more machines are working on it (maybe they come up with validate errors as well): see here

there are more WU like this but mostly (only) two machines with validate errors so far:
http://einstein.phys.uwm.edu/workunit.php?wuid=18982777
http://einstein.phys.uwm.edu/workunit.php?wuid=18980247
http://einstein.phys.uwm.edu/workunit.php?wuid=18994947
http://einstein.phys.uwm.edu/workunit.php?wuid=18640632

and even more where my machines are the only ones with Validate errors so far.

I find this quite bizzare, especially the first one, where all machines seem to agree that the WU is a validate error in itself ;)

btw.. in the first two month I did not experience any validate errors either and now find them sporadically happen on different machines of mine.


____________

Hi, i also got some validate errors on my machines.


machine 1
http://einstein.phys.uwm.edu/workunit.php?wuid=18856427
http://einstein.phys.uwm.edu/workunit.php?wuid=18848997
http://einstein.phys.uwm.edu/workunit.php?wuid=18842557
http://einstein.phys.uwm.edu/workunit.php?wuid=18835952
machine 2
http://einstein.phys.uwm.edu/workunit.php?wuid=18898382
http://einstein.phys.uwm.edu/workunit.php?wuid=18893997
machine 3
http://einstein.phys.uwm.edu/workunit.php?wuid=18936577


All my machines broken????
The problem i think is not by my side... :(
____________

All hosts are getting validating errors on These WU’s.

http://einstein.phys.uwm.edu/workunit.php?wuid=19049072
http://einstein.phys.uwm.edu/workunit.php?wuid=19065962
http://einstein.phys.uwm.edu/workunit.php?wuid=19070317
http://einstein.phys.uwm.edu/workunit.php?wuid=19057172

All this WU’s are from the same data file.
My host is the sixth result on this WU l1_0297.0_S5R1__147_S5R1a.

One other strange thing is that we are a lot of fast hosts doing this data file with short results.

____________
Then you're really interested in a subject, there is no way to avoid it. You have to read the Manual.

Dear Einstein@Home participants,

I'm sorry to say that after more than 150 days of trouble-free project operation, one of our five validator instances went on the rampage at about 10:30 UTC 16-11-2006. I learned about this from the message boards and from an email (thanks Timothy!) around 00:15 UTC 17-11-2006. I restarted the errant validator and all is well again.

My apologies for this glitch - I don't know what went wrong on the server side but experience shows that in spite of our best efforts to run a stable project, these things do sometimes happen.

Credit hounds (and others): I am granting credit for ALL those results which were marked as invalid or validate error during the time window above.

[Subsequent edit/addition]: another validator instance started to show the same problem the next day. I have restarted ALL the validator instances, and granted credit for the second round of problematic results as well.

I apologize again,

Bruce Allen
____________

Thanks a lot for the information.
So we haven't to report here to get our credit because it would be (re)granted on an automated base?
____________

Bruce, thanks for the information and the very good handling of the problem - we can't ask for anything more!

MrS
____________
Scanning for our furry friends since Jan 2002

My apologies for this glitch - I don't know what went wrong on the server side but experience shows that in spite of our best efforts to run a stable project, these things do sometimes happen.

Credit hounds (and others): I am granting credit for ALL those results which were marked as invalid or validate error during the time window above.

I apologize again,

Bruce Allen

You are welcome Dr. Allen, and thank you for looking into the situation and repairing the rogue validator. I believe everyone appreciates the stability of this project and can accept a glitch once every few months, especially since it was such a minor glitch and easily repaired.

____________

You've got infinite numbers in the result files. This usually points to a hardware problem, CPU getting too hot, memory fault etc.

BM

Dear Einstein@Home participants,

I'm sorry to say that after more than 150 days of trouble-free project operation, one of our five validator instances went on the rampage at about 10:30 UTC 16-11-2006. I learned about this from the message boards and from an email (thanks Timothy!) around 00:15 UTC 17-11-2006. I restarted the errant validator and all is well again.

My apologies for this glitch - I don't know what went wrong on the server side but experience shows that in spite of our best efforts to run a stable project, these things do sometimes happen.

Credit hounds (and others): I am granting credit for ALL those results which were marked as invalid or validate error during the time window above.

I apologize again,

Bruce Allen

I just left my compiter witha cleaning programme on it -
when I cam back everything had disappeared!
Then when I logged on here the whole thing had gone down ---
It wan't me was it !!! LOL!!
____________

Dear Einstein@Home participants,

I'm sorry to say ....
<snip>
I apologize again,

Bruce Allen

Please look at the validator again because this one just showed up
http://einstein.phys.uwm.edu/workunit.php?wuid=19025946
Both reported AFTER you thought you solved the problem less than an hour ago Thanks JR

Thankfully a minor problem easily solved.

Having being a fairly keen DC player since 99, I must comment that E@H is remarkably stable. Well done, and your efforts are most appreciated!

FD
____________

Please look at the validator again because this one just showed up
http://einstein.phys.uwm.edu/workunit.php?wuid=19025946
Both reported AFTER you thought you solved the problem less than an hour ago Thanks JR

Please look at the validator again because this one just showed up
http://einstein.phys.uwm.edu/workunit.php?wuid=19025946
Both reported AFTER you thought you solved the problem less than an hour ago Thanks JR

I am betting it was uploaded and validated, then you reported it, which made the validation show up already invalid.

now 2 more

http://einstein.phys.uwm.edu/workunit.php?wuid=19075676
http://einstein.phys.uwm.edu/workunit.php?wuid=19075681
Seems that validator needs more than a reboot...

Here too:

http://einstein.phys.uwm.edu/workunit.php?wuid=18997936
http://einstein.phys.uwm.edu/workunit.php?wuid=18998236
http://einstein.phys.uwm.edu/workunit.php?wuid=18697171

that is not good...
____________

These 4 have been send out, and returned, today. My hosts will do a re-run on them. So I don't know yet, if I get a Validate error as well.

http://einstein.phys.uwm.edu/workunit.php?wuid=19152971
http://einstein.phys.uwm.edu/workunit.php?wuid=19126086
http://einstein.phys.uwm.edu/workunit.php?wuid=19134626
http://einstein.phys.uwm.edu/workunit.php?wuid=19136351

OK, I have restarted all five instances of the validator: one of the OTHER instances was showing problems also. I hope this fixes it. I will modify the credit-granting script to give credit for these results also.

Bruce
____________

OK, I have restarted all five instances of the validator: one of the OTHER instances was showing problems also. I hope this fixes it. I will modify the credit-granting script to give credit for these results also.

Bruce

Mr. allen i cant say it any better than the others here. this is one of the best projects imo some of the others could learn a thing or two from you guys! thank you sir!
____________
seeing without seeing is something the blind learn to do, and seeing beyond vision can be a gift.

Would the credit show up on the web pages? Doesn't seem to be there, e.g.:
http://einstein.phys.uwm.edu/workunit.php?wuid=19065442

Graham.

...Credit hounds (and others): I am granting credit for ALL those results which were marked as invalid or validate error during the time window above...

Would the credit show up on the web pages? Doesn't seem to be there, e.g.:
http://einstein.phys.uwm.edu/workunit.php?wuid=19065442

Graham.

...Credit hounds (and others): I am granting credit for ALL those results which were marked as invalid or validate error during the time window above...

Maybe that needs some times?
I think script must be writen to re-grant the credit lost for false-invalidated WUs otherwise everyone with invalid WUs during the error time would become his credit re-granted.

Would the credit show up on the web pages? Doesn't seem to be there, e.g.:
http://einstein.phys.uwm.edu/workunit.php?wuid=19065442

Graham.

...Credit hounds (and others): I am granting credit for ALL those results which were marked as invalid or validate error during the time window above...

Actually you had ALREADY been granted credit for the result, but the web page 'workunit.php' was not displaying this properly. If you had clicked on the result in question then you would have seen the granted credit on that next web page that would have been displayed.

Anyway I have modified the 'workunit.php' web page so that it now correctly shows the granted credit.

Cheers,
Bruce

Please look at the validator again because this one just showed up
http://einstein.phys.uwm.edu/workunit.php?wuid=19025946
Both reported AFTER you thought you solved the problem less than an hour ago Thanks JR

I am betting it was uploaded and validated, then you reported it, which made the validation show up already invalid.

No Pooh Bear I have a constant connection and use trux client report immediately option....look at my host list I have reported 3 wu's since Bruce thought he fixed the problem

Aha! Many thanks, Bruce.

Graham.

Actually you had ALREADY been granted credit for the result, but the web page 'workunit.php' was not displaying this properly. If you had clicked on the result in question then you would have seen the granted credit on that next web page that would have been displayed.

Anyway I have modified the 'workunit.php' web page so that it now correctly shows the granted credit.

Looks like these (15 to 17 Nov) were the lucky days for hosts like this one, getting huge credit for nothing, with tens of results like this:

Outcome - Client state - CPU time - claimed credit - granted credit
Client error - Compute error - 68.52 - 0.21 - 12.33
Client error - Compute error - 69.58 - 0.19 - 12.33
Client error - Compute error - 72.89 - 0.22 - 12.33

A sample WU (the host in error is the first):
Outcome - Client state - CPU time - claimed credit - granted credit
Client error - Compute error - 72.89 - 0.22 - 12.33
Validate error – Done - 3,273.28 - 12.33 - 12.33
Validate error – Done - 10,131.96 - 12.33 - 12.33
Validate error – Done - 5,065.23 - 12.33 - 12.33
Validate error – Done - 8,220.56 - 12.33 - 12.33
Success – Done - 3,109.99 - 12.33 - 12.33
Success – Done - 2,773.07 - 12.33 - 12.33

If this is due to the validator running in a special mode then it’s fine. But if it is "normal", I can see a problem there. A lot of people would be tempted to use an erratic ultra fast client.