4HTC.bound.mppk.min.pdb_dock_score12_ddg.xml_yfsong_14675_4426_0
1AHW.bound.mppk.min.pdb_dock_score12_ddg.xml_yfsong_14675_4425_0

The above WUs had run times of about 1:35 hours without setting any check points.

They terminated after 99 models complete and uploaded OK (21MB total - ouch!)

.

Something similar.
This WU:
2PTC.bound.mppk.min.pdb_dock_score12_ddg.xml_yfsong_14675_3356_0
ran ongoing for nearly 2 hours (=processor time) when I switched my laptop off.
Today I booted my laptop again and the WU started from 0:00.
Seems to me there was no check-pointing.

Path7.

The next WU (most likely) didn't checkpoint:
1BRS.bound.mppk.min.pdb_dock_score12_ddg.xml_yfsong_14675_3545_0
ran ongoing for 2.5 hours (preference: switch between application every 1 hour has been denied)
Ended by watchdog; 100 decoys.

Path7.

ran ongoing for nearly 2 hours (=processor time) when I switched my laptop off.
Today I booted my laptop again and the WU started from 0:00.
Seems to me there was no check-pointing.

I think I must have had the same problem. I left BOINC running by itself for the last few weeks only to find today that a WU issued on 30th August had only 2.5 hours of processing time (Ncalmodulin_PCS_BOINC_abrelax.1xcycles.v1_SAVE_ALL_OUT_14667_3695_0). Needless to say I missed the report deadline for that one and the second WU I was holding in reserve.

Unfortunately these bugs can happen now and then, but I hope no one else out there has lost 2 weeks of processing time because of it.

---

I am getting jerked on time as well. Almost getting 50 percent or less for time spent crunching. No where close to what seti or others are giving for time spent.

No more crunching rosetta for a while.