• 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.

Assigment Server Connection Problems

mpatc

n00b
Joined
Jun 3, 2004
Messages
32
Can someone please take a look at these logs and tell me what I am doing wrong?

I am trying to update my client on my box at work. I have permission to run the sw but have been told don’t bother the IT support folks with questions or ask for changes (ie firewall changes).

I have been running the ancient Gromacs core v1.90 from March of 2006 since it came out. I tried to update the core sometime last year to something newer but always ran into connection issues. At the time, I said just go back to what works and have been running v1.90 since. My box was recently replaced and I now have a Quad core Xeon and a Quadro FX1700 that I would like to make the best use of. While v1.90 of the CPU core continues to connect and run just fine, I cannot get the 6.23 GPU client to connect and get a WU. I have tinkered with the XP SP3 firewall settings and added each cores .exe but no luck. I have tried to connect to the page that tests your machines connection to the assignment server (page address escapes me at the moment) and it returns OK on port 80 but not on 8080 as I believe that is blocked by our firewall. Doing searches through this forum and the F@H forums, I have tinkered with the config settings to no avail.


When I run the v1.90 client, it looks like this. (showing end of last unit, server connections, start of new unit)



[22:27:43] Writing local files
[22:27:43] Completed 245000 out of 250000 steps (98)
[23:05:36] Writing local files
[23:05:36] Completed 247500 out of 250000 steps (99)
[23:43:38] Writing local files
[23:43:38] Completed 250000 out of 250000 steps (100)
[23:43:38] Writing final coordinates.
[23:43:40] Past main M.D. loop
[23:44:40]
[23:44:40] Finished Work Unit:
[23:44:40] - Reading up to 2304144 from "work/wudata_09.arc": Read 2304144
[23:44:40] - Reading up to 350676 from "work/wudata_09.xtc": Read 350676
[23:44:40] goefile size: 0
[23:44:40] logfile size: 39308
[23:44:40] Leaving Run
[23:44:41] - Writing 2712304 bytes of core data to disk...
[23:44:41] ... Done.
[23:44:42] - Shutting down core
[23:44:42]
[23:44:42] Folding@home Core Shutdown: FINISHED_UNIT
[23:44:44] CoreStatus = 64 (100)
[23:44:44] Sending work to server


[23:44:44] + Attempting to send results
[23:44:51] + Results successfully sent
[23:44:51] Thank you for your contribution to Folding@Home.
[23:44:51] + Number of Units Completed: 299

[23:44:56] - Preparing to get new work unit...
[23:44:56] + Attempting to get work packet
[23:44:56] - Connecting to assignment server
[23:44:57] Couldn't send HTTP request to server (wininet)
[23:44:57] + Could not connect to Assignment Server
[23:44:57] - Successful: assigned to (171.65.103.160).
[23:44:57] + News From Folding@Home: Welcome to Folding@Home
[23:44:57] Loaded queue successfully.
[23:45:08] + Closed connections
[23:45:08]
[23:45:08] + Processing work unit
[23:45:08] Core required: FahCore_78.exe
[23:45:08] Core found.
[23:45:08] Working on Unit 00 [October 6 23:45:08]
[23:45:08] + Working ...
[23:45:08]
[23:45:08] *------------------------------*
[23:45:08] Folding@Home Gromacs Core
[23:45:08] Version 1.90 (March 8, 2006)
[23:45:08]
[23:45:08] Preparing to commence simulation
[23:45:08] - Looking at optimizations...
[23:45:08] - Created dyn
[23:45:08] - Files status OK
[23:45:12] - Expanded 2988150 -> 15111425 (decompressed 505.7 percent)
[23:45:12] - Starting from initial work packet
[23:45:12]
[23:45:12] Project: 2493 (Run 45, Clone 28, Gen 3)
[23:45:12]
[23:45:12] Assembly optimizations on if available.
[23:45:12] Entering M.D.
[23:45:20] Protein: system
[23:45:20]
[23:45:20] Writing local files
[23:45:23] Extra SSE boost OK.
[23:45:24] Writing local files
[23:45:24] Completed 0 out of 250000 steps (0)
[00:23:10] Writing local files
[00:23:10] Completed 2500 out of 250000 steps (1)
[01:01:07] Writing local files
[01:01:07] Completed 5000 out of 250000 steps (2)

When I try the a newer CPU client or the 6.23 GPU client it looks like this. The Machine ID it set to 5 because I have 4 instances of the v1.90 core running, one on each of the CPU cores.


Note: Please read the license agreement (FaH-GPU.exe -license). Further
use of this software requires that you have read and accepted this agreement.



--- Opening Log file [October 8 13:20:00 UTC]


# Windows GPU Console Edition #################################################
###############################################################################

Folding@Home Client Version 6.23

http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\Documents and Settings\username\Desktop\Installs\folding\Folding@home-Win32-GPU_XP-623
Executable: FaH-GPU.exe
Arguments: -advmethods -verbosity 9

[13:20:00] - Ask before connecting: No
[13:20:00] - User name: mpatc (Team 33)
[13:20:00] - User ID: 3C5449544C453E45
[13:20:00] - Machine ID: 5
[13:20:00]
[13:20:00] Loaded queue successfully.
[13:20:00] - Preparing to get new work unit...
[13:20:00] - Autosending finished units... [October 8 13:20:00 UTC]
[13:20:00] + Attempting to get work packet
[13:20:00] Trying to send all finished work units
[13:20:00] + No unsent completed units remaining.
[13:20:00] - Autosend completed
[13:20:00] - Will indicate memory of 3071 MB
[13:20:00] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 7, Stepping: 6
[13:20:00] - Connecting to assignment server
[13:20:00] Connecting to http://assign-GPU.stanford.edu:8080/
[13:20:01] - Couldn't send HTTP request to server
[13:20:01] + Could not connect to Assignment Server
[13:20:01] Connecting to http://assign-GPU.stanford.edu:80/
[13:20:01] Posted data.
[13:20:01] Initial: 543C; + Could not connect to Assignment Server 2
[13:20:01] + Couldn't get work instructions.
[13:20:01] - Attempt #1 to get work failed, and no other work to do.
Waiting before retry.
[13:20:07] + Attempting to get work packet
[13:20:07] - Will indicate memory of 3071 MB
[13:20:07] - Connecting to assignment server
[13:20:07] Connecting to http://assign-GPU.stanford.edu:8080/
[13:20:08] - Couldn't send HTTP request to server
[13:20:08] + Could not connect to Assignment Server
[13:20:08] Connecting to http://assign-GPU.stanford.edu:80/
[13:20:08] Posted data.
[13:20:08] Initial: 543C; + Could not connect to Assignment Server 2
[13:20:08] + Couldn't get work instructions.
[13:20:08] - Attempt #2 to get work failed, and no other work to do.
Waiting before retry.
[13:20:22] + Attempting to get work packet
[13:20:22] - Will indicate memory of 3071 MB
[13:20:22] - Connecting to assignment server
[13:20:22] Connecting to http://assign-GPU.stanford.edu:8080/
[13:20:23] - Couldn't send HTTP request to server
[13:20:23] + Could not connect to Assignment Server
[13:20:23] Connecting to http://assign-GPU.stanford.edu:80/
[13:20:23] Posted data.
[13:20:23] Initial: 543C; + Could not connect to Assignment Server 2
[13:20:23] + Couldn't get work instructions.
[13:20:23] - Attempt #3 to get work failed, and no other work to do.
Waiting before retry.
Any suggestions would be greatly appreciated. I am not sure what changed between the v1.90 clients connectivity and the newer CPU and GPU clients but I cannot seem to get the newer stuff to connect and grab a WU.
 
Try to remove -advmethods and check if it connect fine. Some flags might force the client to pick certain server and some won't respond to requests.
 
I actually added the -advmethods after reading a forum post that said it opened up a larger pool of possible WUs and may help getting a WU assigned. I just tried it without the -advmethods and nothing changed.
 
.............cut................

[23:44:57] Couldn't send HTTP request to server (wininet)

.............................cut...........................

[13:20:07] Connecting to http://assign-GPU.stanford.edu:8080/
and
[13:20:08] Connecting to http://assign-GPU.stanford.edu:80/

..........cut.............

First part:- You don't want to use IE to connect through with the v1.90 client.
Its a legacy setting from the days of dailup.

Second part:- Both those links should come back with ok.
If they don't, then try turning your firewall off and retry them, then trun the firewall back on.
That will show if the firewall is blocking the connection or not.

Luck ............ :D
 
Thanks for the tip on the v1.90 client. I have kept the same settings for years, just copying the folder over to new PCs as I got them. I will get it changed eventually but at least it is working at the moment.

As I said in the original post, Port 8080 is blocked by the main corp firewall. The link on port 80 returns OK but the link on port 8080 returns 'connection was reset'. I have no control over that firewall and have been told not to ask the IT support folks for any help/changes. I also cannot turn off the Windows XP SP3 firewall due to a group policy. I can add exceptions which I have for 80 and 8080 but cannot turn it completely off.

Is port 8080 access absolutely necessary to connect to the assignment server or can it work over port 80 only?
 
You can check which client/server uses which port Here.
I think looking at that table, the GPU client needs port 8080 open to get work.

Luck .......... :D
 
Back
Top