4/27/2006 7:49:18 PM|rosetta@home|Unrecoverable error for result HB_BARCODE_30_1a68__351_24311_2 (aborted via GUI RPC)


5.01 error. Had over 85 hours cpu time and claimed to be just over 11% done. Should've been watching more closely. I wasn't aware there was a problem until I checked the news.

Ackiss

Aborted this workunit ID 12454830. Two other computers failed to return this WU, or any others, so they may *still* be crunching it! See HB_BARCODE_30_1ctf_351_39751

Result ID 18270294

Stopped at 20:05 hours and 8..518%

Run on an HP D530, Pentium 4 @ 2.66 GHz (stock), under WIN XP Home.

---

I thought FACONTACTS_RECENTER* were meant to be removed from the workunit queue on your end? Why are known bad units still being sent out? This one is now back in queue for the next lucky recipient.


name: FACONTACTS_RECENTER_NOFILTERS_1rnbA_448_803_1
WU name: FACONTACTS_RECENTER_NOFILTERS_1rnbA_448_803
project URL: https://boinc.bakerlab.org/rosetta/
report deadline: Thu May 11 13:42:38 2006
app version num: 501
checkpoint CPU time: 38443.460000
current CPU time: 40928.280000
fraction done: 0.016453
VM usage: 0.000000
resident set size: 0.000000
estimated CPU time remaining: 67723.265742
supports graphics: no


https://boinc.bakerlab.org/rosetta/result.php?resultid=18357585

---

[snip]

Please do not abort you current 5.01 Work Units if they are runing well. The science from them is still important to the project. Many of you have asked if credit will be awarded of these failed Work units, and the answer is yes.
[snip]

My current work unit has been running since Monday eve. and shows progress
at 5.90% and CPU time of 148:47:21. Does this meet the criteria of
"running well".

I don't mind letting this process continue to run.....
I'm just worried that its not getting finish anyway....

Mike Hall / Engineering Solutions, Inc.

---

I just realized there was a thread to even report this. I'd have to go through all my results just to see which and when. I'm new to rosetta and figured this out when all my team was having the same issues with the same WU's. Checked previous results on the WU's and no one had finished. I know it was at least 8 WU's for my server host alone. Do I need to pull up all the facts for you?

-edit-
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=14041906
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=14925740
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=14560586
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=14226095
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=14226095
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=13409472
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=13408713
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=13408400
Hope I listed them correctly. Again this is one user. But I guess its their job to report theirs. The ones with the lower cpu time with abort were stuck at 1% for approx 1+hrs. to my recollection.
-edit-

Sincerely,

F. Bulson
Endonet Inc.

---

I lost 44 hours of cpu time, i'm aborting it now it hung @ 1.04%
HBLR_1.0_1ogw_420_6744_2

---

Hi,

have aborted Result ID 18179436 "HBLR_1.0_1b72_ROT_TRIALS_TRIE_449_13_1"
after 10,5 hours at 5%.

---

I lost 44 hours of cpu time, i'm aborting it now it hung @ 1.04%
HBLR_1.0_1ogw_420_6744_2

Can i get my credit for it?
https://boinc.bakerlab.org/rosetta/result.php?resultid=18194769

---

Hi, I just aborted FA_RLXpt_hom003_1ptq__361_16_3 . It was running for more than 12 hours, while my runtime preference is default (4 hours). Also Rhiju suggested in this post that WUs with 1ptq in the title should be aborted. Why I had WU that was told to be wrong like 2 weeks ago?

Going nowhere, the last 5.01 on this box, 38 minutes for 1.06% and a rising completion time:

18432562
Name AB_CASP6_t216__456_5100_0
Workunit 15217912
Created 27 Apr 2006 16:50:33 UTC
Sent 27 Apr 2006 21:04:16 UTC
Received 28 Apr 2006 13:20:57 UTC
Server state Over
Outcome Client error
Client state Computing
Exit status -197 (0xffffff3b)
Computer ID 190714
Report deadline 11 May 2006 21:04:16 UTC
CPU time 2293.640625

Result ID 18229643
Name HBLR_1.0_1hz6_ROT_TRIALS_TRIE_449_43_2
Workunit 14630563
Created 25 Apr 2006 17:03:34 UTC
Sent 25 Apr 2006 17:10:07 UTC
Received 28 Apr 2006 16:24:09 UTC
Server state Over
Outcome Client error
Client state Computing
Exit status -197 (0xffffff3b)
Computer ID 168844
Report deadline 9 May 2006 17:10:07 UTC
CPU time 101762.765625

---

---

Four more 5.01 WUs were aborted this morning

50.1 hrs
https://boinc.bakerlab.org/rosetta/result.php?resultid=18296499
HB_BARCODE_30_5croA_351_21027

51.9 hrs
https://boinc.bakerlab.org/rosetta/result.php?resultid=18296492
HB_BARCODE_30_1a19A_351_28780_3

53.0 hrs
https://boinc.bakerlab.org/rosetta/result.php?resultid=18296362
HBLR_1.0_1dtj_ROT_TRIALS_TRIE_449_27

89.6 hrs
https://boinc.bakerlab.org/rosetta/result.php?resultid=18037119
FA_RLXfn_hom001_1fna__357_63

So, are all of the aborted and stuck WUs being granted 300 points, no matter the computation time?

---

So, are all of the aborted and stuck WUs being granted 300 points, no matter the computation time?

Checked that with my error results.

I think that's indeed what happens, this one has crunched for over 5 days.

Result ID 17773392
Name HBLR_1.0_2tif_420_9913_1
Workunit 13429467
Created 20 Apr 2006 21:56:28 UTC
Sent 21 Apr 2006 4:39:59 UTC
Received 26 Apr 2006 19:26:25 UTC
Server state Over
Outcome Client error
Client state Computing
Exit status -177 (0xffffff4f)
Computer ID 147219
Report deadline 5 May 2006 4:39:59 UTC
CPU time 369442.734375
stderr out

<core_client_version>5.3.12.tx36</core_client_version>
<message>Maximum CPU time exceeded
</message>
<stderr_txt>
# random seed: 1574792
# cpu_run_time_pref: 7200

</stderr_txt>

Validate state Invalid
Claimed credit 1792.64206533905
Granted credit 300
application version 5.01

---

So, are all of the aborted and stuck WUs being granted 300 points, no matter the computation time?

As far as I have been able to determine, the work Units are granted what they claim. If they do not claim any credit (mostly from win 98 machines) them I think they have been getting 30 credits. But I have nothing definitive from the project on this, it is from my own observations, so don't hold the project to these numbers until we hear from them directly.

Result ID 18122796
Name HBLR_1.0_1dtj_420_3452_3
Workunit 13389346
Created 24 Apr 2006 15:15:38 UTC
Sent 24 Apr 2006 20:33:06 UTC
Received 27 Apr 2006 5:24:19 UTC
Server state Over
Outcome Client error
Client state Computing
Exit status -177 (0xffffff4f)
Computer ID 175797
Report deadline 8 May 2006 20:33:06 UTC
CPU time 130784.75
stderr out <core_client_version>5.2.13</core_client_version>
<message>Maximum CPU time exceeded
</message>
<stderr_txt>
# random seed: 1597253
# cpu_run_time_pref: 7200
# random seed: 1597253
# random seed: 1597253

</stderr_txt>


Validate state Invalid
Claimed credit 1165.54456988046
Granted credit 300
application version 5.01

---

Here is another stuck 5.01 WU

https://boinc.bakerlab.org/rosetta/result.php?resultid=18392691

I have since exorcised my farm of the 5.01s and the 5.06s... this one slipped by me though.

Thanks for all you hard work at getting these snags worked out, the R@H team deserves a pat on the back for trying to get this fixed so quickly for us crunchers.

---

2 WUs that I aborted because it takes wayyyy to much time, they didn't hang though.

https://boinc.bakerlab.org/rosetta/result.php?resultid=17827510
FACONTACTS_NOFILTERS_1r69__441_248_1

https://boinc.bakerlab.org/rosetta/result.php?resultid=17773776
HBLR_1.0_2tif_420_9927_1

---

Campeones everywhere!

I have also a WU that is taking WAY to long to complete.
It is progressing though, I would like to see it finished.
It's HB_BARCODE_30_1bm8__351_34196_3
allthough I think it is not going to finish in time.

Is it helpfull for the project to have it progress as much as possible ?

---

Is it helpfull for the project to have it progress as much as possible ?

Alex, in general, yes, it's helpful. In your case other work units are running 10,000 seconds and that one looks like you aborted after 443,000 seconds! 5+ days! Unless you changed your preference to be 4 days... you were more than patient with that one.

I see it was crunched on release 5.01. The newer release has the "watchdog" and it should find WUs such as this and end them much sooner, thus saving you those days of wondering.

So, you did the right things here. You were patient and didn't end it in a sudden panic after 2 hrs and 1 minute, you reported it here, and you're now crunching more WUs. And I believe you will find the current release has resolved problems like this as well, so it shouldn't happen again.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/