Page 1 of 3

Not enough a7 WUs

Posted: Fri Jan 05, 2018 4:03 pm
by Dr. Merkwürdigliebe
There is a deplorable lack of a7 WUs...why is that?

Re: Not enough a7 WUs

Posted: Fri Jan 05, 2018 4:33 pm
by JimboPalmer
Firstly, and this is real: A researcher who is already using a4 on a project will stay on a4 until that project finishes, so as to not have to throw out all previous results. Only all new projects will be a7. (over time this will flip flop and folk will wonder where a4 projects are)

Now some perceived shortages: a7 does not work on VMs, if you are on a VM, you won't get them. (this may have been resolved)
a7 is currently having trouble with 32 bit Windows, if you are running 32 Windows, you will not get a7. (this may someday be resolved)

Re: Not enough a7 WUs

Posted: Fri Jan 05, 2018 5:20 pm
by Nathan_P
I'm processing them all :lol: :lol: currently a 13 second tpf on a 13743!

Seriously - I have no idea, maybe one of the a7 servers is offline?

Re: Not enough a7 WUs

Posted: Fri Jan 05, 2018 5:56 pm
by Dr. Merkwürdigliebe
JimboPalmer wrote: Now some perceived shortages: a7 does not work on VMs, if you are on a VM, you won't get them. (this may have been resolved)
a7 is currently having trouble with 32 bit Windows, if you are running 32 Windows, you will not get a7. (this may someday be resolved)
I'm running Linux x64 (no VM). Xeon E3-1230v3.

Re: Not enough a7 WUs

Posted: Fri Jan 05, 2018 8:58 pm
by bruce
Nathan_P wrote:I'm processing them all :lol: :lol: currently a 13 second tpf on a 13743!

Seriously - I have no idea, maybe one of the a7 servers is offline?
This project is from server 155.247.166.219. I am able to connect to http://155.247.166.219:8080 but not to http://155.247.166.219:80

Maybe something (like maybe a proxy server) is blocking your connection to port 8080?

Under normal circumstances, connections that are attempted to port 80 are redirected to port 8080 on the server, but this redirection is less reliable than the server, itself.

Re: Not enough a7 WUs

Posted: Fri Jan 05, 2018 9:01 pm
by bruce
Nathan_P wrote:I'm processing them all :lol: :lol: currently a 13 second tpf on a 13743!

Seriously - I have no idea, maybe one of the a7 servers is offline?
This project is from server 155.247.166.219. I am able to connect to http://155.247.166.219:8080 but not to http://155.247.166.219:80

@Dr. Merkwürdigliebe:
Maybe something (like maybe a proxy server) is blocking your connection to port 8080 :?:

Under normal circumstances, connections that are attempted to port 80 are redirected to port 8080 on the server, but this redirection is less reliable than the server, itself.

Please post segments FAH's log showing your recent attempts where a a4 WU was downloaded -- plus the first couple of pages of that log.

Re: Not enough a7 WUs

Posted: Fri Jan 05, 2018 9:39 pm
by Dr. Merkwürdigliebe
Thanks for your reply but I'm directly connected to the internet via DSL.

Ironically after posting I received the first a7 WU (13738) for a long time...but they are very rare.

Re: Not enough a7 WUs

Posted: Sun Jan 07, 2018 2:48 pm
by toTOW
My old system that can't run A7 is pretty happy with this situation ... :roll:

Re: Not enough a7 WUs

Posted: Sun Jan 07, 2018 4:53 pm
by Dr. Merkwürdigliebe
My relatively new system that can run A7 is rather unhappy with this situation.

11:27:33:WU01:FS00:0xa4:Version 2.27 (Dec. 15, 2010)

2010...

Re: Not enough a7 WUs

Posted: Mon Jan 08, 2018 3:51 am
by ChristianVirtual
Dr.M. Are you having cpu slot with six or seven threads ? If seven pls. try six as seven is a “higher” prime and not supported for many recent projects (same as 5). slot size should be a multiple of 1,2 and 3. Your assignments of A7 might have been from projects allowing 7 (but that’s a guess, though a small chance it is the reason)

Mind sharing the first 100lines off your log file ?

Re: Not enough a7 WUs

Posted: Mon Jan 08, 2018 5:21 pm
by Dr. Merkwürdigliebe
The setting is and has always been "-1". Xeon E3-1230v3 with four cores, i.e. 8 Threads.

Code: Select all

*********************** Log Started 2018-01-08T15:35:43Z ***********************
15:35:43:************************* Folding@home Client *************************
15:35:43:        Website: http://folding.stanford.edu/
15:35:43:      Copyright: (c) 2009-2016 Stanford University
15:35:43:         Author: Joseph Coffland <joseph@cauldrondevelopment.com>
15:35:43:           Args: --child --lifeline 2159 /etc/fahclient/config.xml --run-as
15:35:43:                 fahclient --pid-file=/var/run/fahclient.pid --daemon
15:35:43:         Config: /etc/fahclient/config.xml
15:35:43:******************************** Build ********************************
15:35:43:        Version: 7.4.16
15:35:43:           Date: Jan 6 2017
15:35:43:           Time: 08:08:33
15:35:43:     Repository: Git
15:35:43:       Revision: e12187cbb0bd6937c067b9749af011374563b7b9
15:35:43:         Branch: master
15:35:43:       Compiler: GNU 4.9.2
15:35:43:        Options: -std=gnu++98 -O3 -funroll-loops -ffast-math -mfpmath=sse
15:35:43:                 -fno-unsafe-math-optimizations -msse2
15:35:43:       Platform: linux2 4.8.0-2-amd64
15:35:43:           Bits: 64
15:35:43:           Mode: Release
15:35:43:******************************* System ********************************
15:35:43:            CPU: Intel(R) Xeon(R) CPU E3-1230 v3 @ 3.30GHz
15:35:43:         CPU ID: GenuineIntel Family 6 Model 60 Stepping 3
15:35:43:           CPUs: 8
15:35:43:         Memory: 15.59GiB
15:35:43:    Free Memory: 14.83GiB
15:35:43:        Threads: POSIX_THREADS
15:35:43:     OS Version: 4.14
15:35:43:    Has Battery: false
15:35:43:     On Battery: false
15:35:43:     UTC Offset: 1
15:35:43:            PID: 2161
15:35:43:            CWD: /var/lib/fahclient
15:35:43:             OS: Linux 4.14.8-041408-generic x86_64
15:35:43:        OS Arch: AMD64
15:35:43:           GPUs: 0
15:35:43:  CUDA Device 0: Platform:0 Device:0 Bus:1 Slot:0 Compute:6.1 Driver:9.1
15:35:43:OpenCL Device 0: Platform:0 Device:0 Bus:1 Slot:0 Compute:1.2 Driver:387.12
15:35:43:***********************************************************************
15:35:43:<config>
15:35:43:  <!-- Client Control -->
15:35:43:  <fold-anon v='true'/>
15:35:43:
15:35:43:  <!-- Folding Slot Configuration -->
15:35:43:  <gpu v='false'/>
15:35:43:
15:35:43:  <!-- Network -->
15:35:43:  <proxy v=':8080'/>
15:35:43:
15:35:43:  <!-- Slot Control -->
15:35:43:  <power v='full'/>
15:35:43:
15:35:43:  <!-- User Information -->
15:35:43:  <passkey v='********************************'/>
15:35:43:  <team v='38412'/>
15:35:43:  <user v='Random_Dude'/>
15:35:43:
15:35:43:  <!-- Folding Slots -->
15:35:43:  <slot id='0' type='CPU'/>
15:35:43:</config>
15:35:43:Switching to user fahclient
15:35:43:Trying to access database...
15:35:43:Successfully acquired database lock
15:35:43:Enabled folding slot 00: READY cpu:8
15:35:43:WU00:FS00:Starting
15:35:43:WU00:FS00:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/fahwebx.stanford.edu/cores/Linux/AMD64/Core_a4.fah/FahCore_a4 -dir 00 -suffix 01 -version 704 -lifeline 2161 -checkpoint 15 -np 8
15:35:43:WU00:FS00:Started FahCore on PID 2171
15:35:43:WU00:FS00:Core PID:2175
15:35:43:WU00:FS00:FahCore 0xa4 started
15:35:44:WU00:FS00:0xa4:
15:35:44:WU00:FS00:0xa4:*------------------------------*
15:35:44:WU00:FS00:0xa4:Folding@Home Gromacs GB Core
15:35:44:WU00:FS00:0xa4:Version 2.27 (Dec. 15, 2010)
15:35:44:WU00:FS00:0xa4:
15:35:44:WU00:FS00:0xa4:Preparing to commence simulation
15:35:44:WU00:FS00:0xa4:- Looking at optimizations...
15:35:44:WU00:FS00:0xa4:- Files status OK
15:35:44:WU00:FS00:0xa4:- Expanded 878665 -> 2053928 (decompressed 233.7 percent)
15:35:44:WU00:FS00:0xa4:Called DecompressByteArray: compressed_data_size=878665 data_size=2053928, decompressed_data_size=2053928 diff=0
15:35:44:WU00:FS00:0xa4:- Digital signature verified
15:35:44:WU00:FS00:0xa4:
15:35:44:WU00:FS00:0xa4:Project: 8636 (Run 2, Clone 161, Gen 39)
15:35:44:WU00:FS00:0xa4:
15:35:44:WU00:FS00:0xa4:Assembly optimizations on if available.
15:35:44:WU00:FS00:0xa4:Entering M.D.
15:35:50:WU00:FS00:0xa4:Using Gromacs checkpoints
15:35:50:WU00:FS00:0xa4:Resuming from checkpoint
15:35:50:WU00:FS00:0xa4:Verified 00/wudata_01.log
15:35:50:WU00:FS00:0xa4:Verified 00/wudata_01.trr
15:35:50:WU00:FS00:0xa4:Verified 00/wudata_01.xtc
15:35:50:WU00:FS00:0xa4:Verified 00/wudata_01.edr
15:35:50:WU00:FS00:0xa4:Completed 808200 out of 1250000 steps  (64%)
15:37:19:WU00:FS00:0xa4:Completed 812500 out of 1250000 steps  (65%)
15:40:19:WU00:FS00:0xa4:Completed 825000 out of 1250000 steps  (66%)
15:43:19:WU00:FS00:0xa4:Completed 837500 out of 1250000 steps  (67%)
15:46:17:WU00:FS00:0xa4:Completed 850000 out of 1250000 steps  (68%)
15:49:16:WU00:FS00:0xa4:Completed 862500 out of 1250000 steps  (69%)
15:52:15:WU00:FS00:0xa4:Completed 875000 out of 1250000 steps  (70%)
15:55:13:WU00:FS00:0xa4:Completed 887500 out of 1250000 steps  (71%)
15:58:11:WU00:FS00:0xa4:Completed 900000 out of 1250000 steps  (72%)
16:01:10:WU00:FS00:0xa4:Completed 912500 out of 1250000 steps  (73%)
16:04:08:WU00:FS00:0xa4:Completed 925000 out of 1250000 steps  (74%)
16:07:07:WU00:FS00:0xa4:Completed 937500 out of 1250000 steps  (75%)
16:10:16:WU00:FS00:0xa4:Completed 950000 out of 1250000 steps  (76%)
16:13:17:WU00:FS00:0xa4:Completed 962500 out of 1250000 steps  (77%)
16:16:24:WU00:FS00:0xa4:Completed 975000 out of 1250000 steps  (78%)
16:19:22:WU00:FS00:0xa4:Completed 987500 out of 1250000 steps  (79%)
16:22:20:WU00:FS00:0xa4:Completed 1000000 out of 1250000 steps  (80%)
16:25:21:WU00:FS00:0xa4:Completed 1012500 out of 1250000 steps  (81%)
16:28:22:WU00:FS00:0xa4:Completed 1025000 out of 1250000 steps  (82%)
16:31:27:WU00:FS00:0xa4:Completed 1037500 out of 1250000 steps  (83%)
16:34:33:WU00:FS00:0xa4:Completed 1050000 out of 1250000 steps  (84%)
16:37:32:WU00:FS00:0xa4:Completed 1062500 out of 1250000 steps  (85%)
16:40:31:WU00:FS00:0xa4:Completed 1075000 out of 1250000 steps  (86%)
16:43:22:WU00:FS00:0xa4:Completed 1087500 out of 1250000 steps  (87%)
16:46:21:WU00:FS00:0xa4:Completed 1100000 out of 1250000 steps  (88%)
16:49:24:WU00:FS00:0xa4:Completed 1112500 out of 1250000 steps  (89%)
16:52:17:WU00:FS00:0xa4:Completed 1125000 out of 1250000 steps  (90%)
16:55:15:WU00:FS00:0xa4:Completed 1137500 out of 1250000 steps  (91%)
16:58:23:WU00:FS00:0xa4:Completed 1150000 out of 1250000 steps  (92%)
17:01:21:WU00:FS00:0xa4:Completed 1162500 out of 1250000 steps  (93%)
17:04:11:WU00:FS00:0xa4:Completed 1175000 out of 1250000 steps  (94%)
17:07:00:WU00:FS00:0xa4:Completed 1187500 out of 1250000 steps  (95%)
17:09:56:WU00:FS00:0xa4:Completed 1200000 out of 1250000 steps  (96%)
17:13:05:WU00:FS00:0xa4:Completed 1212500 out of 1250000 steps  (97%)
17:16:10:WU00:FS00:0xa4:Completed 1225000 out of 1250000 steps  (98%)
17:19:10:WU00:FS00:0xa4:Completed 1237500 out of 1250000 steps  (99%)
17:19:10:WU01:FS00:Connecting to 171.67.108.45:8080
17:19:11:WU01:FS00:Assigned to work server 171.67.108.158
17:19:11:WU01:FS00:Requesting new work unit for slot 00: RUNNING cpu:8 from 171.67.108.158
17:19:11:WU01:FS00:Connecting to 171.67.108.158:8080
17:19:13:WU01:FS00:Downloading 806.27KiB
17:19:16:WU01:FS00:Download complete
17:19:16:WU01:FS00:Received Unit: id:01 state:DOWNLOAD error:NO_ERROR project:9031 run:405 clone:0 gen:684 core:0xa4 unit:0x000002faab436c9e56983491885eab2c

Re: Not enough a7 WUs

Posted: Mon Jan 08, 2018 9:11 pm
by ChristianVirtual
Here goes my theory :-(

Re: Not enough a7 WUs

Posted: Sat Jan 27, 2018 6:36 pm
by SomeStones
Finally, after over a month without seeing an a7, I have seen 3 a7 WUs in the past 24 hours. I have 5 CPU machines and the a7s give me over 50% more points than the a4s.

Re: Not enough a7 WUs

Posted: Wed Feb 28, 2018 8:21 am
by DrBB1
For what it's worth, in the past 10.5 days, my main PC has received 24 a4 WU and 4 a7s. Almost all the WUs have come from 155.247.166.219:8080. I used to get SMPs from a variety of servers, but that seemed to stop several months ago, about the time Temple's servers weren't talking well with Stanford's stat servers (coincidence)? In any case, is there any known reason why the vast majority of my WUs are coming from this single server, and more relevantly, could this "selective assignment" have anything to do with the paucity of a7s many of us have been receiving for several months?

Re: Not enough a7 WUs

Posted: Wed Feb 28, 2018 8:37 am
by Joe_H
The known reason is that currently the Temple servers have most WU's available for CPU processing among all servers with CPU projects. This applies to WU's for both cores. There are new projects for both A4 and A7 in progress, they were announced here on the forum within the last week. Some were from another location besides Temple.