Anyone else got a 1acf protein wu? If so, hows it going?

My longest wu with the 4.77 client has taken 6.3 hours to crunch. This 1acf has already crunched 5 hours and has only just moved from 41.67% to 50% and I've the big chunks to come yet. This ones a real doozy!

---

Wave upon wave of demented avengers march cheerfully out of obscurity into the dream.

I have one of those running for 1h15 on a Pentium D 3.0 now. Only 1% far. :-O A real dragon to slay this 1acf.

---

BOINC.BE: For Belgians who love the smell of glowing red cpu's in the morning
Tutta55's Lair

I've had five 1acf work units so far, and they average almost three and a half hours on my machine...quite a bit longer than the 'typical' work unit, which runs in about two hours. I don't know what 1acf is, but when I see it in my BOINC work queue, I shrug and go take a nap. :)

---

I've had a few of these as well. One of them crashed on my Athlon XP 3000+ after over 2 hours crunching with an error code I have not seen before: -1073741680 (0xc0000090).

No benchmarks ran, no switching (Rosetta is the only project on it), machine not rebooted and it's got 512MB of RAM.

https://boinc.bakerlab.org/rosetta/result.php?resultid=315072

That's the only one that's crashed so far though - I have run others successfully (but slowly), e.g.
https://boinc.bakerlab.org/rosetta/result.php?resultid=296636
https://boinc.bakerlab.org/rosetta/result.php?resultid=266755

But I have another one sitting on my laptop (2.8GHz P4, no HT) that's only just moved to 8.33% after more than an hour.

---

*** Join BOINC@Australia today ***

Hmmmmmmmmm. 2h40 far and still at 1%. I am getting worried now. I wonder if it it makes sense to continue this WU. After all, this is definitely not a slow machine. In any case if it's still at 1% by tonight I will suspend it for the weekend.

---

BOINC.BE: For Belgians who love the smell of glowing red cpu's in the morning
Tutta55's Lair

Looking back over my stats. it appears I've had a few of these - the fastest completed in a just under 2 hours - on my Opteron 144 - the slowest in a little over 4 3/4 hours - also on my Opteron 144.

I currently have one on one of the cores of my Pentium D 840, which is/was showing "CPU time" of 01:39:17, "Progress" of 75.00%, and "To completion" 00:33:05, which means ..... not a lot!

I just checked my results and found that on my P4 3.0G machine:

- 1acf took 12,858.36 secs

- non-1acf (1bmbA) took 7,132.19 secs

---

Ok, after reading this thread

https://boinc.bakerlab.org/rosetta/forum_thread.php?id=78

I decided to restart boinc. Now I'm 16 minutes far and the WU is at 8.33%. Sooo, if the first step at 1% takes unusually long, I guess you can only restart it. Good thing this happened before the weekend.

---

BOINC.BE: For Belgians who love the smell of glowing red cpu's in the morning
Tutta55's Lair

just finished a 1acf wu in 2:21:25

---

--
Florian
www.domplatz1.de

I have completed 10+ 1acaf work units on my computers. no problems

but also noticed that a lot of my recent work units have had client errors for somebody else.

https://boinc.bakerlab.org/rosetta/workunit.php?wuid=167069
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=188726
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=180379

and I noticed that one of my computers (my quad server) does not like working on 1cfyA three client errors in two days.

---

It finished overnight, biggy that one. My machine, (3.2GHz HT P-IV, Windows XP SP2, BOINC 4.25, has an average crunch time of 14,000 seconds with a standard deviation of 3,500 - that 1acf took almost 32,000!

---

Wave upon wave of demented avengers march cheerfully out of obscurity into the dream.

@adrianxw

Wow...never seen one take that long. I've seen the 1acf units take anywhere from 8,000 to 14,000 seconds on my machine (Athlon XP 3000+, 1.5GB RAM, WinXP SP2, BOINC 4.45). I applaude your patience; I would have been trying every trick in the book to "kick-start" it after five hours, and probably would have screwed things up in the process. :)

---

I have WU 1acf__abrelax_33006 and it seems to be stuck on 91.67%. Iknow these type WUs can take longer to do but it is running on an AMD 3000+ running Win2000 and so far it has crunched for 12H 45M. I'd say it's time to dump this WU. Any thoughts or experience with this type WU?

---

I have WU 1acf__abrelax_33006 and it seems to be stuck on 91.67%. Iknow these type WUs can take longer to do but it is running on an AMD 3000+ running Win2000 and so far it has crunched for 12H 45M. I'd say it's time to dump this WU. Any thoughts or experience with this type WU?

I wouldn't dump it yet. Give it a few more hours or check it at the end of the day. If it doesn't complete, try to restart the manager and continue the job.

Thanks, David, I'll try those ideas.

---

I restarted the manager. The time backed off to 2h 35m with the percentage dropping back to 83%. In less than 30 min it has finished. Thanks again, David, for the great advice. In the future, I will not wait so long to try this method.

---

Thanks, David, I'll try those ideas.

i had 2 workunits on my machine, running linux, which have stopped working,
i run other projects too with leave applicatuion in memory on.
boinc was still running, the rosetta-client too, but there was no cpu usage.
after restarting boinc it continued and finished this work units it.

koerni

Be aware that there is one bug with the 4.4x and I think with 4.7x versions ... there are times that a communication thread seems to take 100% cpu. This pops up with projects where they have been off-line for awhile. Like this last bit with CPDN, the BOINC Client sometimes does not seem to want to "let go" and you spin your wheels.

USUALLY cured by stopping and restarting or rebooting. If I suspect i tmay happen agina I will suspend communication with the projtect till it domes cack on link ..

rats thing going ...

---