Folding stats way off

Moderators: Site Moderators, FAHC Science Team

Post Reply
foldinghomealone2
Posts: 148
Joined: Sun Jul 30, 2017 8:40 pm

Folding stats way off

Post by foldinghomealone2 »

Don't know if you've realized but the folding system has a big bug.
It contributed way too many points to some users.

For example Curcoin got 320Mio Points at 10.05, 6am. (and 0 points the three updates before). That makes 80Mio average per update instead of ~50Mio they make normally.
All teams seem to be affected.
MeeLee
Posts: 1375
Joined: Tue Feb 19, 2019 10:16 pm

Re: Folding stats way off

Post by MeeLee »

Seems like the servers were offline, and the accumulated points were rewarded later.
JimboPalmer
Posts: 2573
Joined: Mon Feb 16, 2009 4:12 am
Location: Greenwood MS USA

Re: Folding stats way off

Post by JimboPalmer »

For some time, (while the servers were being updated) F@H has not been awarding the Quick Return Bonus.

They consistently promised that when they finished, the QRB would be calculated for the months it was off.

Now that the servers are all updated, the delinquent QRB was added last evening, it took some hours, but perhaps they are all caught up.

As they updated each server, QRB quit for that server, so if more of your WUs were CPU and your buddies were GPU, the 'lost' QRB points may vary, some folders will get many more than others.

[I am just a donor, and have no inside information]
Tsar of all the Rushers
I tried to remain childlike, all I achieved was childish.
A friend to those who want no friends
foldinghomealone2
Posts: 148
Joined: Sun Jul 30, 2017 8:40 pm

Re: Folding stats way off

Post by foldinghomealone2 »

So you're saying everything is right?
Then look at the stats. Everything is wrong.

Like this dude:
https://folding.extremeoverclocking.com ... =&u=814224
0 WUs but today 19Mio

or:
https://folding.extremeoverclocking.com ... =&u=824569

They really need to fix this...
JimboPalmer
Posts: 2573
Joined: Mon Feb 16, 2009 4:12 am
Location: Greenwood MS USA

Re: Folding stats way off

Post by JimboPalmer »

This IS the fix. Imagine Richiei3 had been doing WUs for months and getting no QRB, just base points. Now at 6 AM they award him/her the QRB points for months of work. There are no new WUs, s/he got credit for those when s/ he did the WU. But s/he is owed almost 19 million lost QRB points. And at 6 AM s/he got them, s/he is now caught up for all s/he was shorted all those months..

I only got 1.1 million QRB points, as I do both CPU folding and GPU folding. My understanding is that GPU server QRBs were broken for a shorter period of time, so those that fold on CPUs will get a higher percentage of back QRB, they have been broken longer.

(in my case, I did 12 WUs while the stats server crunched all those back QRB records, so I got 1.1 million points for 12 WUs)

https://folding.extremeoverclocking.com ... =&u=429597

When F@H finds a past mistake, how do you hope they fix it? They chose to award those who did the work their points. They could have stolen their points from those who crunched all that work, but why?

If F@H were run by Marketing majors instead of Biochemists, we would have gotten clear updates about this fix, but if marketing was in charge, we would be computing Kim Kardashian's popularity, not Proteins.
Tsar of all the Rushers
I tried to remain childlike, all I achieved was childish.
A friend to those who want no friends
foldinghomealone2
Posts: 148
Joined: Sun Jul 30, 2017 8:40 pm

Re: Folding stats way off

Post by foldinghomealone2 »

Ok, now I see what you mean.
Sorry for the confusion.
JimboPalmer
Posts: 2573
Joined: Mon Feb 16, 2009 4:12 am
Location: Greenwood MS USA

Re: Folding stats way off

Post by JimboPalmer »

Now they are confirming that they are crediting back points!
https://foldingathome.org/2019/10/07/bonus-recredit/
Tsar of all the Rushers
I tried to remain childlike, all I achieved was childish.
A friend to those who want no friends
snapshot
Posts: 132
Joined: Thu Apr 09, 2009 7:25 pm
Location: Wiltshire, UK

Re: Folding stats way off

Post by snapshot »

I'm glad I found this thread as I was about to query a team getting 14 million points for zero WUs when it had been inactive for some time. It still seems high to me, given the teams overall output.

But, in that case, where's my re-credit got to? I estimate it was only about 500,000 points but it all helps. (user hurwitt_free, team 157620)
bruce
Posts: 20910
Joined: Thu Nov 29, 2007 10:13 pm
Location: So. Cal.

Re: Folding stats way off

Post by bruce »

I can't answer that question. The re-crediting process is a semi-manual process and is naturally restricted to those with (A) server admin rights and (B) the highest level of trust. There are several steps.
* Identify the server(s) with projects needing a recredit.
* Identify the date-time when the problem started and when it ended.
* grep the applicable server logs for recredit candidates.
* make the necessary changes.

If you can identify a pattern of WUs that were missed in steps 1 and 2, maybe I can convince somebody to look at WUs that didn't show up for corrections. Saying there was a total of 500,000 is not useful information. You MUST provide specific examples from the log files on your computer.
snapshot
Posts: 132
Joined: Thu Apr 09, 2009 7:25 pm
Location: Wiltshire, UK

Re: Folding stats way off

Post by snapshot »

It would have been helpful to know that before the logs were flushed. I doubt there was a pattern as it would only have been four or five WUs. I've found what I think are a couple of instances on one client but the logs on the other aren't backed up and so were lost. I've attached a couple of log fragments that I think hold the required info just in case you can make use of them but I won't hold my breath.....

Code: Select all

13:07:05:WU01:FS01:0x21:Completed 5000000 out of 5000000 steps (100%)
13:07:06:WU01:FS01:0x21:Saving result file logfile_01.txt
13:07:06:WU01:FS01:0x21:Saving result file checkpointState.xml
13:07:06:WU01:FS01:0x21:Saving result file checkpt.crc
13:07:06:WU01:FS01:0x21:Saving result file log.txt
13:07:06:WU01:FS01:0x21:Saving result file positions.xtc
13:07:06:WU01:FS01:0x21:Folding@home Core Shutdown: FINISHED_UNIT
13:07:06:WU01:FS01:FahCore returned: FINISHED_UNIT (100 = 0x64)
13:07:06:WU01:FS01:Sending unit results: id:01 state:SEND error:NO_ERROR project:11720 run:0 clone:1192 gen:436 core:0x21 unit:0x000002608ca304e75b8d8e2c3ba65f63
13:07:06:WU01:FS01:Uploading 7.01MiB to 140.163.4.231
13:07:06:WU01:FS01:Connecting to 140.163.4.231:8080
13:07:06:WU00:FS01:Starting
13:07:06:WU00:FS01:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:\FAHClient\cores/cores.foldingathome.org/Win32/AMD64/NVIDIA/Fermi/Core_21.fah/FahCore_21.exe -dir 00 -suffix 01 -version 705 -lifeline 9456 -checkpoint 15 -gpu-vendor nvidia -opencl-platform 0 -opencl-device 0 -cuda-device 0 -gpu 0
13:07:06:WU00:FS01:Started FahCore on PID 5420
13:07:06:WU00:FS01:Core PID:12564
13:07:06:WU00:FS01:FahCore 0x21 started
13:07:07:WU00:FS01:0x21:*********************** Log Started 2019-07-01T13:07:06Z ***********************
13:07:07:WU00:FS01:0x21:Project: 11728 (Run 0, Clone 104, Gen 311)
13:07:07:WU00:FS01:0x21:Unit: 0x0000018d8ca304e75b95bb16e60948b8
13:07:07:WU00:FS01:0x21:CPU: 0x00000000000000000000000000000000
13:07:07:WU00:FS01:0x21:Machine: 1
13:07:07:WU00:FS01:0x21:Reading tar file core.xml
13:07:07:WU00:FS01:0x21:Reading tar file integrator.xml
13:07:07:WU00:FS01:0x21:Reading tar file state.xml
13:07:07:WU00:FS01:0x21:Reading tar file system.xml
13:07:07:WU00:FS01:0x21:Digital signatures verified
13:07:07:WU00:FS01:0x21:Folding@home GPU Core21 Folding@home Core
13:07:07:WU00:FS01:0x21:Version 0.0.20
13:07:10:WU00:FS01:0x21:Completed 0 out of 5000000 steps (0%)
13:07:10:WU00:FS01:0x21:Temperature control disabled. Requirements: single Nvidia GPU, tmax must be < 110 and twait >= 900
13:07:12:WU01:FS01:Upload 24.98%
13:07:18:WU01:FS01:Upload 81.19%
13:07:22:WU01:FS01:Upload complete
13:07:22:WU01:FS01:Server responded WORK_ACK (400)
13:07:22:WU01:FS01:Final credit estimate, 22198.00 points
13:07:22:WU01:FS01:Cleaning up

Code: Select all

15:16:32:WU00:FS01:0x21:Completed 12500000 out of 12500000 steps (100%)
15:16:34:WU00:FS01:0x21:Saving result file logfile_01.txt
15:16:34:WU00:FS01:0x21:Saving result file checkpointState.xml
15:16:34:WU00:FS01:0x21:Saving result file checkpt.crc
15:16:34:WU00:FS01:0x21:Saving result file log.txt
15:16:34:WU00:FS01:0x21:Saving result file positions.xtc
15:16:35:WU00:FS01:0x21:Folding@home Core Shutdown: FINISHED_UNIT
15:16:36:WU00:FS01:FahCore returned: FINISHED_UNIT (100 = 0x64)
15:16:36:WU00:FS01:Sending unit results: id:00 state:SEND error:NO_ERROR project:14180 run:13 clone:470 gen:12 core:0x21 unit:0x0000000f0002894c5d3b569bf264ee93
15:16:36:WU00:FS01:Uploading 8.99MiB to 155.247.166.220
15:16:36:WU00:FS01:Connecting to 155.247.166.220:8080
15:16:36:WU01:FS01:Starting
15:16:36:WU01:FS01:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:\FAHClient\cores/cores.foldingathome.org/Win32/AMD64/NVIDIA/Fermi/Core_21.fah/FahCore_21.exe -dir 01 -suffix 01 -version 705 -lifeline 11704 -checkpoint 15 -gpu-vendor nvidia -opencl-platform 0 -opencl-device 0 -cuda-device 0 -gpu 0
15:16:36:WU01:FS01:Started FahCore on PID 7500
15:16:36:WU01:FS01:Core PID:15712
15:16:36:WU01:FS01:FahCore 0x21 started
15:16:36:WU01:FS01:0x21:*********************** Log Started 2019-08-19T15:16:36Z ***********************
15:16:36:WU01:FS01:0x21:Project: 11714 (Run 0, Clone 3812, Gen 3)
15:16:36:WU01:FS01:0x21:Unit: 0x000000058ca304f15b21ad245de6f9b2
15:16:36:WU01:FS01:0x21:CPU: 0x00000000000000000000000000000000
15:16:36:WU01:FS01:0x21:Machine: 1
15:16:36:WU01:FS01:0x21:Reading tar file core.xml
15:16:36:WU01:FS01:0x21:Reading tar file integrator.xml
15:16:36:WU01:FS01:0x21:Reading tar file state.xml
15:16:36:WU01:FS01:0x21:Reading tar file system.xml
15:16:36:WU01:FS01:0x21:Digital signatures verified
15:16:36:WU01:FS01:0x21:Folding@home GPU Core21 Folding@home Core
15:16:36:WU01:FS01:0x21:Version 0.0.20
15:16:39:WARNING:WU00:FS01:WorkServer connection failed on port 8080 trying 80
15:16:39:WU00:FS01:Connecting to 155.247.166.220:80
15:16:40:WU01:FS01:0x21:Completed 0 out of 7500000 steps (0%)
15:16:40:WU01:FS01:0x21:Temperature control disabled. Requirements: single Nvidia GPU, tmax must be < 110 and twait >= 900
15:16:42:WARNING:WU00:FS01:Exception: Failed to send results to work server: Failed to connect to 155.247.166.220:80: No connection could be made because the target machine actively refused it.
15:16:42:WU00:FS01:Trying to send results to collection server
15:16:42:WU00:FS01:Uploading 8.99MiB to 128.252.203.4
15:16:42:WU00:FS01:Connecting to 128.252.203.4:8080
15:16:48:WU00:FS01:Upload 22.95%
15:16:54:WU00:FS01:Upload 92.51%
15:16:56:WU00:FS01:Upload complete
15:16:56:WU00:FS01:Server responded WORK_ACK (400)
15:16:56:WU00:FS01:Final credit estimate, 29000.00 points
15:16:56:WU00:FS01:Cleaning up
bruce
Posts: 20910
Joined: Thu Nov 29, 2007 10:13 pm
Location: So. Cal.

Re: Folding stats way off

Post by bruce »

FAh normally keeps about a dozen previous logs unless you delete them. (The number can be adjusted.)
Post Reply