Page 1 of 1

128.143.199.96 choking on upload [SMP on classic server?]

Posted: Tue Mar 22, 2011 3:26 pm
by ThunderRd
I don't know why, but recently I have been receiving SMP WUs from this server, which the system labels as "classic", and maintained by Peter. It hasn't been a problem, WUs come and go. Today, a finished WU upped and the download stalled. Several restarts of the client yielded the same result. I downloaded the latest client [I had been using 6.30] and updated it to 6.34 with the same results. It stalls each time as you can see in the log.

I checked the server status page; the server is showing status "accept" which indicates the following: "When a server is in "accept" mode, it will accept WUs, but not give any out. Some servers are used for internal testing of F@H and might seem unfamiliar. "

Before anyone tries to suggest the basics, let me say that this machine has been running for over 4 years and has turned out many thousands of SMP units. No changes have been made to its configuration.

command line:

Code: Select all

"D:\Program Files\smp\Folding@home-Win32-x86.exe" -smp -advmethods -verbosity 9
log:

Code: Select all

Note: Please read the license agreement (Folding@home-Win32-x86.exe -license). F
urther
use of this software requires that you have read and accepted this agreement.

4 cores detected
'mpiexec' is not recognized as an internal or external command,
operable program or batch file.


--- Opening Log file [March 22 15:07:39 UTC]


# Windows SMP Console Edition #################################################
###############################################################################

                       Folding@Home Client Version 6.34

                          http://folding.stanford.edu

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

Launch directory: d:\Program Files\smp
Executable: D:\Program Files\smp\Folding@home-Win32-x86.exe
Arguments: -smp -advmethods -verbosity 9

[15:07:39] - Ask before connecting: No
[15:07:39] - User name: ThunderRd (Team 45)
[15:07:39] - User ID: 1F80DD70569D6EA5
[15:07:39] - Machine ID: 1
[15:07:39]
[15:07:40] Loaded queue successfully.
[15:07:40] - Preparing to get new work unit...
[15:07:40] - Autosending finished units... [March 22 15:07:40 UTC]
[15:07:40] Cleaning up work directory
[15:07:40] Trying to send all finished work units
[15:07:40] + Attempting to get work packet
[15:07:40] + No unsent completed units remaining.
[15:07:40] Passkey found
[15:07:40] - Autosend completed
[15:07:40] - Will indicate memory of 4095 MB
[15:07:40] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 7, Stepping: 6
[15:07:40] - Connecting to assignment server
[15:07:40] Connecting to http://assign.stanford.edu:8080/
[15:07:41] Posted data.
[15:07:41] Initial: 8F80; - Successful: assigned to (128.143.199.96).
[15:07:41] + News From Folding@Home: Welcome to Folding@Home
[15:07:41] Loaded queue successfully.
[15:07:41] Sent data
[15:07:41] Connecting to http://128.143.199.96:8080/
[15:08:02] - Couldn't send HTTP request to server
[15:08:02] + Could not connect to Work Server
[15:08:02] - Attempt #1  to get work failed, and no other work to do.
Waiting before retry.
[15:08:22] + Attempting to get work packet
[15:08:22] Passkey found
[15:08:22] - Will indicate memory of 4095 MB
[15:08:22] - Connecting to assignment server
[15:08:22] Connecting to http://assign.stanford.edu:8080/
[15:08:23] Posted data.
[15:08:23] Initial: 8F80; - Successful: assigned to (128.143.199.96).
[15:08:23] + News From Folding@Home: Welcome to Folding@Home
[15:08:23] Loaded queue successfully.
[15:08:23] Sent data
[15:08:23] Connecting to http://128.143.199.96:8080/
[15:08:35] Posted data.
[15:08:35] Initial: 0000; - Receiving payload (expected size: 1773522)
[15:10:07] + Could not get Work unit data from Work Server
[15:10:07] - Attempt #2  to get work failed, and no other work to do.
Waiting before retry.
[15:10:18] + Attempting to get work packet
[15:10:18] Passkey found
[15:10:18] - Will indicate memory of 4095 MB
[15:10:18] - Connecting to assignment server
[15:10:18] Connecting to http://assign.stanford.edu:8080/
[15:10:19] Posted data.
[15:10:19] Initial: 8F80; - Successful: assigned to (128.143.199.96).
[15:10:19] + News From Folding@Home: Welcome to Folding@Home
[15:10:20] Loaded queue successfully.
[15:10:20] Sent data
[15:10:20] Connecting to http://128.143.199.96:8080/
[15:10:40] Posted data.
[15:10:40] Initial: 0000; - Receiving payload (expected size: 1773522)
EDIT: Interesting. Shortly after posting this, I had a brain fart and forced a core a3 download. The client immediately jumped onto server 171.64.65.54 and downloaded a project 6020 WU, so for now, the box is crunching on that one. I do wonder why the client has been connecting to the other server, though. Here is an example:

Code: Select all

[21:22:33] + Attempting to send results [March 21 21:22:33 UTC]

[21:22:33] - Reading file work/wuresults_04.dat from core

[21:22:33]   (Read 3530424 bytes from disk)

[21:22:33] Connecting to http://128.143.199.96:8080/

[21:23:11] Posted data.

[21:23:11] Initial: 0000; - Uploaded at ~90 kB/s

[21:23:11] - Averaged speed for that direction ~90 kB/s

[21:23:11] + Results successfully sent

[21:23:11] Thank you for your contribution to Folding@Home.

[21:23:11] + Number of Units Completed: 208


[21:23:15] Trying to send all finished work units

[21:23:15] + No unsent completed units remaining.

[21:23:15] - Preparing to get new work unit...

[21:23:15] Cleaning up work directory

[21:23:15] + Attempting to get work packet

[21:23:15] Passkey found

[21:23:15] - Will indicate memory of 4095 MB

[21:23:15] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 7, Stepping: 6

[21:23:15] - Connecting to assignment server

[21:23:15] Connecting to http://assign.stanford.edu:8080/

[21:23:17] Posted data.

[21:23:17] Initial: 8F80; - Successful: assigned to (128.143.199.96).
 <++++++++++++++++++++++++note this server
[21:23:17] + News From Folding@Home: Welcome to Folding@Home

[21:23:17] Loaded queue successfully.

[21:23:17] Sent data

[21:23:17] Connecting to http://128.143.199.96:8080/
   <++++++++++++++++++++++++note this server
[21:23:19] Posted data.

[21:23:19] Initial: 0000; - Receiving payload (expected size: 1770233)

[21:23:29] - Downloaded at ~172 kB/s

[21:23:29] - Averaged speed for that direction ~165 kB/s

[21:23:29] + Received work.

[21:23:29] Trying to send all finished work units

[21:23:29] + No unsent completed units remaining.

[21:23:29] + Closed connections

[21:23:29] 

[21:23:29] + Processing work unit

[21:23:29] Core required: FahCore_a3.exe

[21:23:29] Core found.

[21:23:29] Working on queue slot 05 [March 21 21:23:29 UTC]

[21:23:29] + Working ...

[21:23:29] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 05 -np 4 -nocpulock -checkpoint 5 -verbose -lifeline 3652 -version 630'


[21:23:29] 

[21:23:29] *------------------------------*

[21:23:29] Folding@Home Gromacs SMP Core

[21:23:29] Version 2.27 (Dec. 15, 2010)

[21:23:29] 

[21:23:29] Preparing to commence simulation

[21:23:29] - Looking at optimizations...

[21:23:29] - Created dyn

[21:23:29] - Files status OK

[21:23:30] - Expanded 1769721 -> 1957708 (decompressed 110.6 percent)

[21:23:30] Called DecompressByteArray: compressed_data_size=1769721 data_size=1957708, decompressed_data_size=1957708 diff=0

[21:23:30] - Digital signature verified

[21:23:30] 

[21:23:30] Project: 6955 (Run 0, Clone 87, Gen 1)

[21:23:30] 

[21:23:30] Assembly optimizations on if available.

[21:23:30] Entering M.D.

[21:23:36] Mapping NT from 4 to 4 

[21:23:36] Completed 0 out of 500000 steps  (0%)

[21:28:53] Completed 5000 out of 500000 steps  (1%)

[21:34:07] Completed 10000 out of 500000 steps  (2%)

[21:39:19] Completed 15000 out of 500000 steps  (3%)

[21:44:30] Completed 20000 out of 500000 steps  (4%)

[21:49:41] Completed 25000 out of 500000 steps  (5%)

[21:54:51] Completed 30000 out of 500000 steps  (6%)

[22:00:02] Completed 35000 out of 500000 steps  (7%)

[22:05:13] Completed 40000 out of 500000 steps  (8%)

[22:10:23] Completed 45000 out of 500000 steps  (9%)

Re: 128.143.199.96 choking on upload [SMP on classic server?

Posted: Tue Mar 22, 2011 6:07 pm
by gwildperson
For a while, SMP servers were unique and only managed SMP projects. I've noticed that newer servers seem to manage both SMP and Uniprocessor projects so my guess is that the designation of certain servers as SMP is going away and soon they'll all be called classic. I've seen no trend to combine GPU servers with classic servers, but since we're rapidly moving toward a unified client (apparently that's what V7 is) maybe we're also migrating toward a unified server. If so, the second column of serverstat.html may disappear.

I can't shed any light on your problem, though.

Re: 128.143.199.96 choking on upload [SMP on classic server?

Posted: Tue Mar 22, 2011 7:46 pm
by 7im
The new A4 core has blurred the lines of what is classic and what is SMP. It can run as either 1 core or multi-core. And like Gdub said, V7 will probably do more blurrrring.