What is up with this work unit

The first two computers went past the deadline so it was issued to two more. Then one of the first two returned a result after the deadline.

Of the second two computers one returned the result before the deadline but the other missed the deadline so it was issued to another computer. Then the result was returned by the one that missed the deadline.

The third re-issue timed out and was re-issued to me.

Computer 4669410 has a validate error and granted credit ---.

Computer 2224122 has "Completed, validation inconclusive" (no concensus) and credit is pending.

Computer 4137258 had "completed and validated" but granted credit zero.

If credit was granted, why was it re-issued to me when computer 4663654 timed out?

Was the science valid or did the validator mess up?

If the science is valid I'll just abort it. However, if credit has been "granted", will the validator look at my result?

Inputs?

What is up with this work unit

The first two computers went past the deadline so it was issued to two more. Then one of the first two returned a result after the deadline.

Of the second two computers one returned the result before the deadline but the other missed the deadline so it was issued to another computer. Then the result was returned by the one that missed the deadline.

The third re-issue timed out and was re-issued to me.

Computer 4669410 has a validate error and granted credit ---.

Computer 2224122 has "Completed, validation inconclusive" (no concensus) and credit is pending.

Computer 4137258 had "completed and validated" but granted credit zero.

If credit was granted, why was it re-issued to me when computer 4663654 timed out?

Was the science valid or did the validator mess up?

If the science is valid I'll just abort it. However, if credit has been "granted", will the validator look at my result?

Inputs?

Looks to me like a validation mess up!

Computer 4669410 has a validate error and granted credit ---.

Has not received credit for the scan. The task ID page shows it as invalid and has granted 0 credit.

Not sure what has happened to be honest.

---

That is...a very strange WU. Someone should get the attention of one of the admins on that one.. strange.

---

Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)

This is a weird one. I've PMed Dr. Korpela to take a look at this thread.

---

Looks like there's bug in the validator that marked an unvalidated result as validated. Fortunately the workunit itself isn't considered validated until a canonical result is identified.

I'm looking for the source of the problem.

---

@SETIEric@qoto.org (Mastodon)

Looks like there's bug in the validator that marked an unvalidated result as validated. Fortunately the workunit itself isn't considered validated until a canonical result is identified.

I'm looking for the source of the problem.

I'll go ahead and process the workunit.

OK, I processed the work unit and it validated against the computer that had been granted the zero credit. We both were granted credit and there is now a canonical result. Thanks for the interest in this.

Those poor two guys/gals who had the validate error/marked as invalid.

---

"Life is a tragedy for those who feel, and a comedy for those who think."

"I never get into an argument that I cannot win."

I too am having issues with the "Completed, validation inconclusive" error. Anyhelp will be appreciated.

Also some trouble with a "client detached" error.

Host 3903055

Thanks..

---

Boinc Wiki




"Great spirits have always encountered violent opposition from mediocre minds." -Albert Einstein

"Completed, Validation inconclusive" is not an error per se, it is simply the validator letting you know that your result did not match closely enough to that of your quorum partner's so a third must be sent out to find which one is canonical.

The "client detached" errors I'm not entirely certain about. Have you been hitting the "reset project" button, or have you been detaching and re-attaching to the project trying to get more work?

---

Hi, Thanks for the response OzzFan,

Problems all started when I checked this machine 1 morning back at the beginning of March.. checked the "Tasks" for their status like every other day and found that everything was marked: "Computation Error" Sadly there were probably 60 WUs. After that I could not get WUs to download and this PC sat idle for about 3 days or so.

I was reading the forums and at 1 point read to detach and reatach and then re-apply the location prefs... so I did. Helped a bit, machine downloaded about 20 WUs and thats it. It would not DL anymore until the next day and then another 20 WUs. Anyways, likely that is my "client detached" error right there. I am surprised that it took 2 weeks for that to clear itself up. Oh well.. we all know the systems have been very busy.

Here is a link to 1 of the failed WUs:

WU 1179033360

and here is a link to 1 of the "Completed, Validation inconclusive" WUs.

WU 1173439548

I will wait for a few more days to see what happens with the validating...

---

Boinc Wiki




"Great spirits have always encountered violent opposition from mediocre minds." -Albert Einstein

Hi, Thanks for the response OzzFan,

Problems all started when I checked this machine 1 morning back at the beginning of March.. checked the "Tasks" for their status like every other day and found that everything was marked: "Computation Error" Sadly there were probably 60 WUs. After that I could not get WUs to download and this PC sat idle for about 3 days or so.

I was reading the forums and at 1 point read to detach and reatach and then re-apply the location prefs... so I did. Helped a bit, machine downloaded about 20 WUs and thats it. It would not DL anymore until the next day and then another 20 WUs. Anyways, likely that is my "client detached" error right there. I am surprised that it took 2 weeks for that to clear itself up. Oh well.. we all know the systems have been very busy.

Here is a link to 1 of the failed WUs:

WU 1179033360

Hmm, I think the "too many exit(0)s" on those was from Raistmer's team mod, at least task 1182105168 indicates you were using that package. I don't have a CUDA capable GPU so haven't followed the features of that package closely, but IIRC its method of trying to get work to the GPU is to do an exit 0, then when BOINC restarts the task it should end up using CUDA. There don't seem to be such errors since March 10, I'd advise not worrying about those unless they start happening again.

and here is a link to 1 of the "Completed, Validation inconclusive" WUs.

WU 1173439548

I will wait for a few more days to see what happens with the validating...

That's a very interesting case where your CPU result with 4 spikes found was compared against 2 CUDA results, one of which got an early result_overflow with 31 pulses but the other found 4 spikes. I expect you'll get credit when the 4th host reports, when work is computed on different hardware there's a higher likelihood of minor computational differences adding up to an imperfect match but it usually only causes one of the signals to be mismatched. Again, unless you get a lot of those you probably shouldn't worry, and I didn't see any in your recently reported work.


Edit: I almost forgot, the recent 'Client detached' indications are from 21 Mar 2009 22:07:38 UTC, so not related to the much earlier case you noted. There have been occasional reports of that happening spontaneously with no user action, and the servers were dealing with recovery from the science database problem at that time. The tasks will have been sent to other hosts, and that's not counted as a host error.

                                                              Joe

Heh.. well I had a big long post set but then hit preview and decided to follow 1 of the links.. oops.. clicked back and the page had expired... lol

anyways.. seems when I click the links (BSs and FreeDC) next to the profile, the wrong computer is being listed too.

wow I really screwed this 1 up.. yet it continues to crunch away..

I will keep an eye on it.. hopefully things clear themselves up soon.. :P

Thanks for the help.

---

Boinc Wiki




"Great spirits have always encountered violent opposition from mediocre minds." -Albert Einstein