FAHClient V7.1.52 released (1st Public-Release)

[smoking2000]

FAHCoreWrapper v7.1.52 for Ubuntu 10.10 64bit (tar.bz2 package) doesn't work on Debian Unstable anymore. It fails after starting the FahCore with: FahCore returned: FAILED_2 (1 = 0x1)

Code: Select all

18:45:57:Trying to access database...
18:45:57:Successfully acquired database lock
18:45:57:Enabled folding slot 00: READY smp:4
18:45:57:Started thread 3 on PID 23768
18:45:57:Started thread 5 on PID 23768
18:45:57:WU00:FS00:Starting
18:45:57:Started thread 4 on PID 23768
18:45:57:Started thread 6 on PID 23768
18:45:57:Started thread 7 on PID 23768
18:45:57:WU00:FS00:Running FahCore: /home/bas/software/folding-fah7/FAHCoreWrapper /home/bas/software/folding-fah7/cores/www.stanford.edu/~pande/Linux/AMD64/Core_a4.fah/FahCore_a4 -dir 00 -suffix 01 -version 701 -lifeline 23768 -checkpoint 5 -np 4 -forceasm -verbose -nice 19
18:45:57:WU00:FS00:Started FahCore on PID 23776
18:45:57:Started thread 8 on PID 23768
18:45:57:WU00:FS00:FahCore 0xa4 started
18:45:57:WU00:FS00:FahCore returned: FAILED_2 (1 = 0x1)
18:45:57:WU00:FS00:Starting
18:45:57:WU00:FS00:Running FahCore: /home/bas/software/folding-fah7/FAHCoreWrapper /home/bas/software/folding-fah7/cores/www.stanford.edu/~pande/Linux/AMD64/Core_a4.fah/FahCore_a4 -dir 00 -suffix 01 -version 701 -lifeline 23768 -checkpoint 5 -np 4 -forceasm -verbose -nice 19
18:45:57:WU00:FS00:Started FahCore on PID 23779
18:45:57:Started thread 9 on PID 23768
18:45:57:WU00:FS00:FahCore 0xa4 started
18:45:57:WU00:FS00:FahCore returned: FAILED_2 (1 = 0x1)
18:46:57:WU00:FS00:Starting
18:46:57:WU00:FS00:Running FahCore: /home/bas/software/folding-fah7/FAHCoreWrapper /home/bas/software/folding-fah7/cores/www.stanford.edu/~pande/Linux/AMD64/Core_a4.fah/FahCore_a4 -dir 00 -suffix 01 -version 701 -lifeline 23768 -checkpoint 5 -np 4 -forceasm -verbose -nice 19
18:46:57:WU00:FS00:Started FahCore on PID 23818
18:46:57:Started thread 10 on PID 23768
18:46:57:WU00:FS00:FahCore 0xa4 started
18:46:57:WU00:FS00:FahCore returned: FAILED_2 (1 = 0x1)
18:46:57:WARNING:WU00:FS00:Too many errors, failing
18:46:57:WU00:FS00:Sending unit results: id:00 state:SEND error:FAILED project:7610 run:626 clone:0 gen:81 core:0xa4 unit:0x0000006d664f2dd04de6d4f5f949689b
18:46:57:WU00:FS00:Connecting to 171.64.65.104:8080
18:46:57:WU01:FS00:Connecting to assign3.stanford.edu:8080
18:46:58:WU00:FS00:Server responded WORK_QUIT (404)
18:46:58:WARNING:WU00:FS00:Server did not like results, dumping
18:46:58:WU00:FS00:Cleaning up
18:46:58:WU01:FS00:News: Welcome to Folding@Home
18:46:58:WU01:FS00:Assigned to work server 171.64.65.104
18:46:58:WU01:FS00:Requesting new work unit for slot 00: READY smp:4 from 171.64.65.104
18:46:58:WU01:FS00:Connecting to 171.64.65.104:8080
18:46:59:WU01:FS00:Downloading 329.92KiB
18:47:00:WU01:FS00:Download complete
18:47:00:WU01:FS00:Received Unit: id:01 state:DOWNLOAD error:OK project:7611 run:2 clone:2 gen:154 core:0xa4 unit:0x000000ae664f2dd04df0f55783bfe329
18:47:00:WU01:FS00:Starting
18:47:00:WU01:FS00:Running FahCore: /home/bas/software/folding-fah7/FAHCoreWrapper /home/bas/software/folding-fah7/cores/www.stanford.edu/~pande/Linux/AMD64/Core_a4.fah/FahCore_a4 -dir 01 -suffix 01 -version 701 -lifeline 23768 -checkpoint 5 -np 4 -forceasm -verbose -nice 19
18:47:00:WU01:FS00:Started FahCore on PID 23823
18:47:00:Started thread 11 on PID 23768
18:47:00:WU01:FS00:FahCore 0xa4 started
18:47:01:WU01:FS00:FahCore returned: FAILED_2 (1 = 0x1)
18:47:01:WU01:FS00:Starting
18:47:01:WU01:FS00:Running FahCore: /home/bas/software/folding-fah7/FAHCoreWrapper /home/bas/software/folding-fah7/cores/www.stanford.edu/~pande/Linux/AMD64/Core_a4.fah/FahCore_a4 -dir 01 -suffix 01 -version 701 -lifeline 23768 -checkpoint 5 -np 4 -forceasm -verbose -nice 19
18:47:01:WU01:FS00:Started FahCore on PID 23826
18:47:01:Started thread 12 on PID 23768
18:47:01:WU01:FS00:FahCore 0xa4 started
18:47:01:WU01:FS00:FahCore returned: FAILED_2 (1 = 0x1)
^C18:47:03:WARNING:Caught signal SIGINT(2) on PID 23768
18:47:03:Exiting, please wait. . .
18:47:04:Clean exit


When using FAHCoreWrapper from v7.1.50 with FAHClient v7.1.52 it works correctly:

Code: Select all

18:48:40:Trying to access database...
18:48:40:Successfully acquired database lock
18:48:40:Enabled folding slot 00: READY smp:4
18:48:40:Started thread 3 on PID 23898
18:48:40:Started thread 4 on PID 23898
18:48:40:Started thread 5 on PID 23898
18:48:40:WU01:FS00:Starting
18:48:40:Started thread 6 on PID 23898
18:48:40:Started thread 7 on PID 23898
18:48:40:WU01:FS00:Running FahCore: /home/bas/software/folding-fah7/FAHCoreWrapper /home/bas/software/folding-fah7/cores/www.stanford.edu/~pande/Linux/AMD64/Core_a4.fah/FahCore_a4 -dir 01 -suffix 01 -version 701 -lifeline 23898 -checkpoint 5 -np 4 -forceasm -verbose -nice 19
18:48:40:WU01:FS00:Started FahCore on PID 23906
18:48:40:Started thread 8 on PID 23898
18:48:40:WU01:FS00:Core PID:23910
18:48:40:WU01:FS00:FahCore 0xa4 started
18:48:41:WU01:FS00:0xa4:
18:48:41:WU01:FS00:0xa4:*------------------------------*
18:48:41:WU01:FS00:0xa4:Folding@Home Gromacs GB Core
18:48:41:WU01:FS00:0xa4:Version 2.27 (Dec. 15, 2010)
18:48:41:WU01:FS00:0xa4:
18:48:41:WU01:FS00:0xa4:Preparing to commence simulation
18:48:41:WU01:FS00:0xa4:- Assembly optimizations manually forced on.
18:48:41:WU01:FS00:0xa4:- Not checking prior termination.
18:48:41:WU01:FS00:0xa4:- Expanded 337328 -> 644556 (decompressed 191.0 percent)
18:48:41:WU01:FS00:0xa4:Called DecompressByteArray: compressed_data_size=337328 data_size=644556, decompressed_data_size=644556 diff=0
18:48:41:WU01:FS00:0xa4:- Digital signature verified
18:48:41:WU01:FS00:0xa4:
18:48:41:WU01:FS00:0xa4:Project: 7611 (Run 2, Clone 2, Gen 154)
18:48:41:WU01:FS00:0xa4:
18:48:41:WU01:FS00:0xa4:Assembly optimizations on if available.
18:48:41:WU01:FS00:0xa4:Entering M.D.
18:48:47:WU01:FS00:0xa4:Completed 0 out of 2000000 steps  (0%)


[smoking2000]

FAHCoreWrapper v7.1.52 for Ubuntu 10.10 64bit (tar.bz2 package) doesn't work on Debian Unstable anymore. It fails after starting the FahCore with: FahCore returned: FAILED_2 (1 = 0x1)
[...]
When using FAHCoreWrapper from v7.1.50 with FAHClient v7.1.52 it works correctly:

Using the FAHCoreWrapper from v7.1.52 Ubuntu 10.10 64bit deb package also works correctly.

Why are the files in .tar.bz2 release not the same as in the package?

[Jesse_V]

Screen-lag is another problem for those who have not participated in the previous betas. Folks need a clear warning about that, and once it's posted to the main page with the default configuration of GPU=True we'll be getting lots of complaints. You can blame the FAHCore or the driver for problems with GPUs, but it's V7 that's configuring the GPU by default, and that's the real problem for a public release that enables cores and drivers that aren't ready for novices.

The warning needs to be on the download page.

Exactly. Putting it on the Download page is an excellent idea, but of course we don't want to scare people away from V7 itself. Another options is show the message on V7's configuration screen. When V7 first starts up and doesn't see an existing configuration, it asks for a slot configuration and has SMP+GPU as the default option if the hardware supports it. This is fine for science, but perhaps the GPU slot should be optional UNTIL the cores handle their processing a bit better and the problem goes away. Of course this would require J. Coffland to make some coding tweaks, and I know he's very busy to get this out there. From my perspective, I'm not entirely certain how common lag issues actually are. All I know is that on both my laptop's 240m and my desktop's 560TI, GPU folding makes the most system impact, but on both machines the issues are fairly tolerable. The v6 Nvidia GPU client caused some display issues on my parent's machines, which may have been heat related in retrospect. Thus in my mind installing SMP/uniprocessor by default is fine, but exactly like you so eloquently said I'm also not sure about the widespread release of the GPU folding if my experiences are in fact the norm.

[RLHay]

Here is a preview of the new front page: http://folding.stanford.edu/English/Main2

That new main page is a great improvement - well done to all involved.

[Zagen30]

I do like the new home page, though I agree with elrado1 that the periods on Ready. Set. Fold. are unnecessary. I know that passkeys are mentioned in the v7 installation guide, but I feel like there should be some mention of it in the Earn points, join a team window; if you're going to be competitive, you usually need a passkey.

[Spitfire19]

I'll probably download it on April 1st providing no major problems arise.

[Spitfire19]

I agree there should be a warning about the GPU if you are going to install it as default. Some of the newer GPU Wu's do have the potential to over tax and possibly damage a highly OCed and inadequately cooled GPU and they will also make your cpu sluggish if a person is unaware of how they run and the problems that might arise there could be some resulting damage. If I was a new folder and installed v7 F@H on my gaming rig which has a ASUS Matrix GTX 580 on it and when it is gaming I run it at an OC of 1000+, if I try to fold at that with the fans set to auto it will hit close to the 100C mark. That is something that a person really should be warned about and be aware of.

Yep, Always be around watching GPU-Z and Speedfan while doing the first couple WUs.

[bruce]

Thanks for your report, csimon. Nobody knows what causes this problem (yet). I'm going to move your question to a topic we already have going on the issue, but I'll leave this quote here so it's still reported against V7.1.52. See viewtopic.php?f=67&t=20566

I'm still getting the following error - Microsoft Visual C++ Runtime Library

"Assertion failed!

Program: C:\Program Files (x86)\FAHClient\FAHControl.exe
File: cairo-surface.c
Line: 860

Expression: status==CAIRO_STATUS_SUCCESS

For information on how your program can cause an assertion
failure, see the Visual C++ documentation on asserts

(Press Retry to debug the application - JIT must be enabled)"

It only happens if I install V7 for both gpu&smp ...
My workaround has been to use V7 for gpu & V6 for smp.
I tried today to run both together and it failed again with the same message.

So, what info do I need to provide and, how do I fix this?

[Spitfire19]

Screen-lag is another problem for those who have not participated in the previous betas. Folks need a clear warning about that, and once it's posted to the main page with the default configuration of GPU=True we'll be getting lots of complaints. You can blame the FAHCore or the driver for problems with GPUs, but it's V7 that's configuring the GPU by default, and that's the real problem for a public release that enables cores and drivers that aren't ready for novices.

The warning needs to be on the download page.

Personally I feel that the default setting should be Uniprocessor when initially installing the system. I always set it up as Uni first when trying it on new machines. Get some monitoring tools up before dedicating the machine to F@H.

[965@57*C]
[6950@64*C]
On 7.1.50

[7im]

New donors are not likely to want to setup FAH 2 or 3 times like you do.

But I agree that we could do better guiding new donors to the appropriate setup options the first time through.

[csimon]

Thanks for your report, csimon. Nobody knows what causes this problem (yet). I'm going to move your question to a topic we already have going on the issue, but I'll leave this quote here so it's still reported against V7.1.52. See viewtopic.php?f=67&t=20566

thanks bruce

[Jesse_V]

And the site is live! Wow major changes everywhere! Thanks J. Coffland, I know you're making a lot of the changes!
On the Press FAQ page (http://folding.stanford.edu/English/FAQ-Press) all apostrophes are rendered as strange triangle symbols. Please fix.
V7's installation guide still has Beta Client Warnings despite being linked to from the main page, which seems contradictory to me. Is it still necessary? Maybe for the GPU slot...

Also, http://folding.typepad.com/news/2012/03 ... sease.html

[7im]

You seem to have some misconceptions about what a public release means. Wasn't private beta vs. public beta explained a while ago? What does the title of this thread say?

[HendricksSA]

Let me say "Great job!". The v7 client is something you can all be proud of and the site is looking much more 21st century.

[jsfitz54]

New 7.1.52 GPU running hotter than 7.1.50 @ 86-88c . And that's at 100% fan speed!
I eliminated the GPU slot, again.