• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Early Unit End?

S

sobe

Limp Gawd
Joined
Feb 23, 2006
Messages
346
Anyone else notice any Early Unit Ends on their boxes? I have never had one before, and now I've had 3 on 3 different boxes, 2 of which are brand new stock clocked Dell's

Code: 
[15:00:09] *------------------------------*
[15:00:09] Folding@Home Gromacs Core
[15:00:09] Version 1.90 (March 8, 2006)
[15:00:09] 
[15:00:09] Preparing to commence simulation
[15:00:09] - Looking at optimizations...
[15:00:09] - Files status OK
[15:00:13] - Expanded 3067643 -> 16583069 (decompressed 540.5 percent)
[15:00:15] 
[15:00:15] Project: 2414 (Run 108, Clone 22, Gen 12)
[15:00:15] 
[15:00:22] Assembly optimizations on if available.
[15:00:22] Entering M.D.
[15:00:45] (Starting from checkpoint)
[15:00:45] Protein: p2414_Ribo_tryptophan280
[15:00:45] 
[15:00:45] Writing local files
[15:00:45] Completed 152500 out of 250000 steps  (61)
[15:00:49] Extra SSE boost OK.
[15:42:19] Writing local files
[15:42:19] Completed 155000 out of 250000 steps  (62)
[16:20:46] Writing local files
[16:20:46] Completed 157500 out of 250000 steps  (63)
[16:52:46] Gromacs cannot continue further.
[16:52:46] Going to send back what have done.
[16:52:46] logfile size: 103838
[16:52:46] - Writing 104374 bytes of core data to disk...
[16:52:46]   ... Done.
[16:52:46] 
[16:52:46] Folding@home Core Shutdown: EARLY_UNIT_END
[16:52:50] CoreStatus = 72 (114)
[16:52:50] Sending work to server
 
[18:43:12] Project: 2413 (Run 0, Clone 16, Gen 9)
[18:43:12]
[18:43:13] Assembly optimizations on if available.
[18:43:13] Entering M.D.
[18:43:21] Protein: p2413_Ribo_isoleucine
[18:43:21]
[18:43:21] Writing local files
[18:43:26] Extra SSE boost OK.
[18:43:27] Writing local files
[18:43:27] Completed 0 out of 250000 steps (0)
[19:19:37] Writing local files
[19:19:37] Completed 2500 out of 250000 steps (1)
[19:56:41] Writing local files
[19:56:41] Completed 5000 out of 250000 steps (2)
[20:32:55] Writing local files
[20:32:55] Completed 7500 out of 250000 steps (3)
[21:09:52] Writing local files
[21:09:52] Completed 10000 out of 250000 steps (4)
[21:46:31] Writing local files
[21:46:31] Completed 12500 out of 250000 steps (5)
[22:23:14] Writing local files
[22:23:14] Completed 15000 out of 250000 steps (6)
[22:40:41] Gromacs cannot continue further.
[22:40:41] Going to send back what have done.
[22:40:41] logfile size: 17294
[22:40:41] - Writing 17830 bytes of core data to disk...
[22:40:41] ... Done.
[22:40:41]
[22:40:41] Folding@home Core Shutdown: EARLY_UNIT_END
[22:40:45] CoreStatus = 72 (114)
[22:40:45] Sending work to server
 
ND40oz said:
Are you still getting points?
Click to expand...

Do you have any idea how impossible that is for me to tell? :D

Yea, you get partial credit for EUE's.
 
ND40oz said:
Are you still getting points?
Click to expand...

Yes, but only 53 or 59 pts... a far cry from the 600 points I would have gotten if it had finished... not to mention the loss of science..
 
sobe said:
Yes, but only 53 or 59 pts... a far cry from the 600 points I would have gotten if it had finished... not to mention the loss of science..
Click to expand...

Not necessarily a loss to science, they may already have confirmed the answer and don't need the rest of the work unit...
 
ND40oz said:
Not necessarily a loss to science, they may already have confirmed the answer and don't need the rest of the work unit...
Click to expand...

I did not know that a unit ended early for any other reason than an unexpected error, such as a miscalculation by the CPU
 
sobe said:
I did not know that a unit ended early for any other reason than an unexpected error, such as a miscalculation by the CPU
Click to expand...

From what I've seen, the "Gromacs cannot continue further" message is not a cpu error but the fact that something turned up wrong with the folding of that gromac. There is no need to do any further work on it and it ends early. You get credited for the number of steps completed divided by the total number of steps times the point value for the work unit.
 
A certian number of EUE's are a normal part of the program.
 
I've got 54 GPU units completed on the 1900xtx and the only time I get EUE
is after I get core outdated
 
ND40oz said:
From what I've seen, the "Gromacs cannot continue further" message is not a cpu error but the fact that something turned up wrong with the folding of that gromac. There is no need to do any further work on it and it ends early. You get credited for the number of steps completed divided by the total number of steps times the point value for the work unit.
Click to expand...

I dont doubt what you're saying about the reason why it ended early, but the points dont add up. By your calculation I should have gotten 378 points out of the 600 for the whole WU but I only recieved 59 points. This is killing my point production, I havent had a gromac finish completley in a couple of days. I think I will disable the large work units untill this stops happening.
 
If it has happened to several WU in a row, the first think I would do is delete the FAH core which will force download of current core. This has fixed similar problems for me before.

RPhArrow --\\\------------------------->
 
Got one myself today



[22:39:55] Writing local files
[22:39:55] Completed 145000 out of 500000 steps (29)
[22:41:40] Writing checkpoint files
[22:54:07] Writing local files
[22:54:07] Completed 150000 out of 500000 steps (30)
[22:54:07] Going to send back what have done.
[22:54:07] logfile size: 27014
[22:54:07] - Writing 27534 bytes of core data to disk...
[22:54:07] ... Done.
[22:54:07]
[22:54:07] Folding@home Core Shutdown: EARLY_UNIT_END
[22:54:10] CoreStatus = 72 (114)
[22:54:10] Sending work to server


this was [15:33:46] Protein: p1809_Collagen_POG10new_refolding core
 
sobe said:
I havent had a gromac finish completley in a couple of days.
Click to expand...

When did last check your box is stable ?
If overclocked then set to default speed.

I run about 1 early end per 100 work units folded.
But they do tend to come in batches.
So once a week I will run through all my log files a check the status of each work unit as it finished.
If a box has more than 1 early end then it gets a quick health check.
Temps & volts, 2 loops of memtest, reinstall the core, etc, etc.

Luck ............. :D
 
Tigerbiten said:
When did last check your box is stable ?
If overclocked then set to default speed.

I run about 1 early end per 100 work units folded.
But they do tend to come in batches.
So once a week I will run through all my log files a check the status of each work unit as it finished.
If a box has more than 1 early end then it gets a quick health check.
Temps & volts, 2 loops of memtest, reinstall the core, etc, etc.
Click to expand...

The machines that have had the early end's are not overclocked, 2 are brand new Dell GX520's, and another is a 3 month old Dell Precision D380, the last one is my home machine which is a homebrew AMD X2 3800+ that has been OC'd in the past, but is running stock clock right now. I read over at the folding community forums that there has been an increase in EUE's as of late, so I think it may not be related to my hardware.
 
You must log in or register to reply here.
Back
Top