Folding again

I started Folding again Wednesday. I see the guys at Stanford still haven't done any work on the client. I have 3 PCs folding 24/7 and it has crashed all 3 at least once. It even crashed my main PC that hasn't crashed in 9 months of running 24/7. It crashed my laptop for the first time

Still doesn't properly utilize a P4 with HT but I have found that Windows Task Manager may be reading the CPU usage wrong. My Asus Ai Booster is reading CPU 0 at 90% and CPU 1 at 50% for an average of 70%. I can live with that.

PC 1
P4C 3.2GHz @ 3.5GHz 512KB cache (Northwood) W/ HT 1024MB PC3200 DDR
Finished 1 WU Wed and due to finish another one 02:14AM Fri 12 May 06

PC 2
P4A 2.0GHz 512KB cache (Northwood) W/O HT 768MB PC2700 DDR
Due to finish first WU at 15:53 Fri 12 May 06

PC 3 HP/Compaq NX9600 laptop
P4 640 3.2GHz 2048KB cache (Prescott) W/ HT 1024MB PC3200 DDR
Due to finish first WU at 02:49 Fri 12 May 06

 

I'm convinced something is not right - see the PC's in my sig? They're all folding 24/7 and my daily average is around 300 points...

 

If I get the three WUs that are running finished they should give me 585 more points and move me up 100 slots in the team ranks.

 

I think I lose WUs.
I have 6 PCs running 24/7 and I register as having 4 CPUs sending in WUs in the last 7 Days and a mid 300s daily point average.
No 2 of my PCs take more then a week to do a project so I cant see how I can be only showing 2 less then I am running.
I even checked how they are set up to insure the WUs are not going elsewhere.

 

CPUs are only counted as running if they turn a finished WU. If a CPU looses a WU or can't send a finished WU and hasn't turned in a finished WU it is not counted as running.

All 3 of my CPUs have turned in a finished WU so it shows up as having 3 CPUs but I have lost 3 WUs in 2 days because of crashes. I turned off the send and receive WU larger than 5MB so I may be able to finish some between crashes. I was getting 400 point WUs that take 3 days to finish but now I'm getting 50 point WUs that take 4-5 hours. 2 of my CPUs still have large WUs to finish I hope they don't crash before they are done and can start on the smaller WUs.

 

I do understand that but I used to register all CPUs and WUs atleast once a week, this is telling me my computer is not finishing a WU in less then 7 days on two boxes and I know it not to be true. I check them and it shows no more then 3 days until the WU is to be finished, 4 days later....nothing.

 

You could be loosing WU before they are finished. I have lost them at 95% after 3 days of running fine.

[23:22:39] Completed 11750000 out of 12500000 steps (94)
[00:01:01] Writing local files
[00:01:01] Completed 11875000 out of 12500000 steps (95)
[00:23:43] Quit 101 - Fatal error:
[00:23:43] Step 11948932, time 23897.9 (ps) LINCS WARNING
[00:23:43] relative constraint deviation after LINCS:
[00:23:43] max 0.000565 (between atoms 213 and 215) rms 0.000080
[00:23:43]
[00:23:43] Simulation instability has been encountered. The run has entered a
[00:23:43] state from which no further progress can be made.
[00:23:43] This may be the correct result of the simulation, however if you
[00:23:43] often see other project units terminating early like this
[00:23:43] too, you may wish to check the stability of your computer (issues
[00:23:43] such as high temperature, overclocking, etc.).
[00:23:43] Going to send back what have done.
[00:23:43] logfile size: 300966
[00:23:43] - Writing 301650 bytes of core data to disk...
[00:23:43] ... Done.
[00:23:43]
[00:23:43] Folding@home Core Shutdown: EARLY_UNIT_END
[00:23:46] CoreStatus = 72 (114)
[00:23:46] Sending work to server


[00:23:46] + Attempting to send results
[00:23:56] + Results successfully sent
[00:23:56] Thank you for your contribution to Folding@Home.
[00:24:00] - Preparing to get new work unit...
[00:24:00] + Attempting to get work packet
[00:24:00] - Connecting to assignment server
[00:24:00] - Successful: assigned to (171.65.103.151).
[00:24:00] + News From Folding@Home: Welcome to Folding@Home
[00:24:00] Loaded queue successfully.
[00:24:01] + Closed connections
[00:24:06]
[00:24:06] + Processing work unit
[00:24:06] Core required: FahCore_79.exe
[00:24:06] Core found.
[00:24:06] Working on Unit 05 [May 19 00:24:06]
[00:24:06] + Working ...
[00:24:07]
[00:24:07] *------------------------------*
[00:24:07] Folding@Home Double Gromacs Core
[00:24:07] Version 1.91 (April 11, 2006)
[00:24:07]
[00:24:07] Preparing to commence simulation
[00:24:07] - Looking at optimizations...
[00:24:07] - Created dyn
[00:24:07] - Files status OK
[00:24:07] - Expanded 154626 -> 629661 (decompressed 407.2 percent)
[00:24:07] - Starting from initial work packet
[00:24:07]
[00:24:07] Project: 991 (Run 5200, Clone 0, Gen 17)
[00:24:07]
[00:24:07] Assembly optimizations on if available.
[00:24:07] Entering M.D.
[00:24:13] Protein: p991_longfkligand
[00:24:13]
[00:24:13] Writing local files
[00:24:13] Extra SSE2 boost OK.
[00:24:14] Writing local files
[00:24:14] Completed 0 out of 500000 steps (0)

Three days wasted, and 172 points lost, with only 3.5 hours left to go.

When something like that happens you don't get credit for the WU or a CPU running


The PC didn't crash it just started a new WU. The new WU will take 1 1/2 days for 99 points

 

Have you tried changing the client.cfg machineid?