I can report simlar problems on 2 WUs. One is reporting 7 hour elapsed, 26% progress and 24 hours to completion. The 2nd reports 16 hours elapsed, 11% progress and 51 hours to completion!!!! The progress is going up, but so is the time to completion.

Hi Jean, and thanks for the info - I'll watch and calculate and see if we are getting closer or further to completion.
The WU that seems 'normal' is BETA_CMD2_0001-DHRS3.clustersOccur-MYH2A.clustersOccur-137_1 and the WU that seems to be a'monster' is BETA_CMD2_0001-1RKC_A.clustersOccur-1YDI_A.clustersOccur_86_2 They seem to be from the same 'family'.
The 'monster' above is now running at high priority, and is showing 16 hours elapsed 13% progress and 51 Hours to completion.
The 'normal' Wu is running on an Intel Pentium M 1700MHz in an IBM Thinkpad, and the 'monster' is running on an Intel Celeron 3.06GHz in a Packard Bell desktop system

Processing BETA_CMD2_0001-PP1BA.clustersOccur-TPM3A.clustersOccur_4_1
Currently 10.191% after 6:40:26
Estimated 16:50:52 remaining, but simple math takes me to 52 hours remaining

Xeon R5404 (2Ghz), also running another Beta (BETA_CMD2_0001-PP1BA.clustersOccur-VINCA.clustersOccur_5_1 21% after 2 hours, set to finish in 10) and 2 energy work units, not sure if its affecting anything at all.

Hi all,
I picked a monster one too:

BETA_ CMD2_ 0001-1RKC_ A.clustersOccur-1YDI_ A.clustersOccur_ 10_ 0--


This WU are running in high priority in the last 26 hours now.

The time to completion are more than 34 hours now, but continue increasing each more.

The increment of work done are happens in very slow motion, but are growing anyway (never saw this kind of speed for any WU in more than 4 years of WCG... including many Betas WU's crunched before...).

There is no special resources drenage happens until now in my machine ( a INTEL QUAD EXTREME QX9770 running at 3.2 GHz, W Vista 32 Ultimate , 4 GB RAM, crunching more 3 HFCC WU's at sametime, at 100% of processor dedication to WCG). Just monitored for around half hour and saw that this monster WU are using around 5.400 kB of memory to crunch (very little if compared with 115.000 KB in average of any HFCC WU's that I'm crunching at sametime, for instance).

One from my WU quorum just give up to continue crunch, and abort your WU after 19 hours. None are finished yet.

I'm monitoring this one each 3 horus, once means that are crunching without issues, despite of the lack of speed, to avoid bad surprises.

I hope that it finishing in good shape in one or two days, despite my little concern about the completion deadline scheduled to 4/27 12:08 PM.

Checkpoint at 10 % for my BETA_ CMD2_ 0001-PP1BA.clustersOccur-TPM1A.clustersOccur_ 34_ 0

   Runtime        % done    Extrapd Total Runtime
  2.5 hours       6.00 %       42 hours
  8.5 hours       8.56 %      100 hours
~10 hours         9.25 %      110 hours
 11.89 hours     10.002 %     118.87 hours

Extrapolated total runtime is computed from runtime and % already done. Boinc's time to completion is useless in these circumstances (~17 hours right now!).

It seems more and more obvious that this exponential WU might never reach 100 %. I will let it live a little longer in case I have some good idea or one of the techs tells us what to do. And then I will probably have to kill it. I'll let you know.

Beside this seriously annoying problem everything looks fine and in line with other WUs for this new project:
RAM use: 5,400 kB
Peak memory use: 5,572 kB
VM size: 30,212 kB
Total page faults: 18,336 (for 12.5 hours!)
Other WUs running in the other cores are not affected at all.
And checkpointing occurs as frequently as usual for this project.
Biggest checkpoint file in the slot directory: 169 kB. All others at 1 kB.

Cheers. Jean.