Linux SMP client hangs at start [at Entering M.D.]

[Oak37]

Hi,
Recently one of my machines running the SMP client stopped working, when I went to check it the client had appeared to have hung just after Entering M.D. It started doing this straight after it uploaded a work unit a few days ago.
So far I have installed a new copy of the client into a different folder and ran it without the -smp and -oneunit flag, it got a unit ran through it and uploaded it fine. When I started it with the -smp flag it hung again. I can't really think of what else to do, any ideas?
Here are my system details:
Ubuntu 7.10
Core2duo e 6420 @ 2.511 GHz
1Gb RAM

Here's the most recent part of the log file, it stops at Entering M.D.
It doesn't show on the log file but when I run the client in the terminal it outputs 4 lines onto the end:
NNODES=4, MYRANK=0, HOSTNAME=Apollo
NNODES=4, MYRANK=1, HOSTNAME=Apollo
NNODES=4, MYRANK=2, HOSTNAME=Apollo
NNODES=4, MYRANK=2, HOSTNAME=Apollo

Code: Select all

--- Opening Log file [March 26 17:02:55]


# SMP Client ##################################################################
###############################################################################

                       Folding@Home Client Version 6.01beta2

                          http://folding.stanford.edu

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

Launch directory: /home/david/NewFolding
Executable: ./fah6
Arguments: -oneunit -forceasm -smp

Warning:
By using the -forceasm flag, you are overriding
safeguards in the program. If you did not intend to
do this, please restart the program without -forceasm.
If work units are not completing fully (and particularly
if your machine is overclocked), then please discontinue
use of the flag.

[17:02:55] - Ask before connecting: No
[17:02:55] - User name: Oak37 (Team 52286)
[17:02:55] - User ID: 4271EC983C62FB91
[17:02:55] - Machine ID: 5
[17:02:55]
[17:02:55] Loaded queue successfully.
[17:02:55]
[17:02:55] + Processing work unit
[17:02:55] Core required: FahCore_a1.exe
[17:02:55] Core found.
[17:02:55] Working on Unit 02 [March 26 17:02:55]
[17:02:55] + Working ...
[17:02:55]
[17:02:55] *------------------------------*
[17:02:55] Folding@Home Gromacs SMP Core
[17:02:55] Version 1.74 (November 27, 2006)
[17:02:55]
[17:02:55] Preparing to commence simulation
[17:02:55] - Ensuring status. Please wait.
[17:03:12] - Assembly optimizations manually forced on.
[17:03:12] - Not checking prior termination.
[17:03:13] - Expanded 2433161 -> 12894661 (decompressed 529.9 percent)
[17:03:13] - Starting from initial work packet
[17:03:13]
[17:03:13] Project: 2653 (Run 22, Clone 77, Gen 83)
[17:03:13]
[17:03:13] Assembly optimizations on if available.
[17:03:13] Entering M.D.

Thanks!

[7im]

Is there an entry in the etc/hosts file for Apollo? What's it say?

[Oak37]

Is there an entry in the etc/hosts file for Apollo? What's it say?

Thanks for replying!

I've got:
127.0.0.1 localhost
127.0.1.1 Apollo

... the rest is just things about IPv6.

[7im]

127.0.1.1 Apollo

What happens if you change that to 127.0.0.1 ?

[Oak37]

What happens if you change that to 127.0.0.1 ?

Nope, nothing I restarted the PC just to make sure and upon restart the file had saved the changes I made but the client is still hanging.

I had recently installed citadel server before this issue arose, it didn't affect the work unit that was running at the time. I uninstalled it though, made sure to remove any entries pertaining to it as well so don't know if that could be the issue.

[7im]

No more light bulbs going on at the moment. And no other posts about citadel. Sorry.

Anyone else?

[Oak37]

No more light bulbs going on at the moment. And no other posts about citadel. Sorry.

Anyone else?

No problem, thanks for the help
I'm a bit lost as where to even start with it, it works without the smp flag which makes it even stranger

[él Mero]

Just checked my hosts-file, replace:

Code: Select all

127.0.0.1 localhost

with:

Code: Select all

127.0.0.1 localhost.localdomain localhost Apollo

Don't know if you need to delete:

Code: Select all

127.0.1.1 Apollo

But delete it and try, and if it doesn't work put it back and try again.

[Oak37]

Thanks for that el mero,
I swapped all of them around but to no avail, I also checked the hosts file for my two other Ubuntu machines and they have the same format as listed on my broken one, Apollo.
I've googled it and there was quite a bit of info, all of them were related to mismatched hostnames and nearly all of them were solved simply by redoing the hosts file, not to sure what I'm doing wrong. I guess I'll go through all different permeatations.

[dschief]

Are you still having trouble?

did you try running ifconfig
to look at packet transmission status
it will show data for eth0 and lo

you could try

service network stop

service network start

to reinitialize hardware.

[Oak37]

Sorry for the late reply, I haven't had much time to try and fix it although I feel I've exhausted all the different options available in the hosts file. At the moment I'm just going to wait until the next version of Ubuntu is released and update all my servers and folders from scratch, hopefully a clean start will do the trick albeit in an extreme way.
Thanks for everyone's help, much appreciated

[So_it_goes99]

Hi

Since upgrading to FAH6.01beta1 in February I'v had the same problem.
The folding hangs at Entering MD.
Can't run SMP clients but 2 'normal' clients runs OK.

The latest FAH6.02beta1 gives the same problems.

Tried the /etc/hosts 'fix'. but nothing helps.

//

[7im]

Hello So_it_goes99, welcome to the forum.

Please post the first 30 lines of your fahlog.txt to help us diagnose the problem. Please also include an example of the hang, if not in that first 30 lines.

Thanks.

[So_it_goes99]

Hi

Thanks for the reply.
I had a clean system downloaded the client
and estarted everything from scratch.

[user@host smp]$ ./fah6 -smp -verbosity 9 &
[1] 7979
[user@host smp]$
Note: Please read the license agreement (fah6 -license). Further
use of this software requires that you have read and accepted this agreement.

2 cores detected


--- Opening Log file [April 10 14:27:05]


# SMP Client ##################################################################
###############################################################################

Folding@Home Client Version 6.02beta

http://folding.stanford.edu

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

Launch directory: /home/user/folding/smp
Executable: ./fah6
Arguments: -smp -verbosity 9

[14:27:05] - Ask before connecting: No
[14:27:05] - User name: So_it_goes (Team 37451)
[14:27:05] - User ID not found locally
[14:27:05] + Requesting User ID from server
[14:27:05] - Getting ID from AS:
[14:27:05] Connecting to http://assign.stanford.edu:8080/
[14:27:06] Posted data.
[14:27:06] Initial: 6E29; - Received User ID = 296E35431B018BDA
[14:27:06] - Machine ID: 3
[14:27:06]
[14:27:06] Work directory not found. Creating...
[14:27:06] Could not open work queue, generating new queue...
[14:27:06] - Autosending finished units...
[14:27:06] Trying to send all finished work units
[14:27:06] + No unsent completed units remaining.
[14:27:06] - Autosend completed
[14:27:06] - Preparing to get new work unit...
[14:27:06] + Attempting to get work packet
[14:27:06] - Will indicate memory of 2000 MB
[14:27:06] - Detect CPU. Vendor: AuthenticAMD, Family: 15, Model: 3, Stepping: 2
[14:27:06] - Connecting to assignment server
[14:27:06] Connecting to http://assign.stanford.edu:8080/
[14:27:06] Posted data.
[14:27:06] Initial: 40AB; - Successful: assigned to (171.64.65.56).
[14:27:06] + News From Folding@Home: Welcome to Folding@Home
[14:27:07] Loaded queue successfully.
[14:27:07] Connecting to http://171.64.65.56:8080/
[14:27:10] Posted data.
[14:27:10] Initial: 0000; - Receiving payload (expected size: 2440327)
[14:27:15] - Downloaded at ~476 kB/s
[14:27:15] - Averaged speed for that direction ~476 kB/s
[14:27:15] + Received work.
[14:27:15] + Closed connections
[14:27:15]
[14:27:15] + Processing work unit
[14:27:15] Core required: FahCore_a1.exe
[14:27:15] Core not found.
[14:27:15] - Core is not present or corrupted.
[14:27:15] - Attempting to download new core...
[14:27:15] + Downloading new core: FahCore_a1.exe
[14:27:15] Downloading core (/~pande/Linux/x86//Core_a1.fah from http://www.stanford.edu)
...
... DOWNLOADING CORE.
...
[14:27:21] Verifying core Core_a1.fah...
[14:27:21] Signature is VALID
[14:27:21]
[14:27:21] Trying to unzip core FahCore_a1.exe
[14:27:21] Decompressed FahCore_a1.exe (3625104 bytes) successfully
[14:27:21] + Core successfully engaged
[14:27:26]
[14:27:26] + Processing work unit
[14:27:26] Core required: FahCore_a1.exe
[14:27:26] Core found.
[14:27:26] Working on Unit 01 [April 10 14:27:26]
[14:27:26] + Working ...
[14:27:26] - Calling './mpiexec -np 4 -host 127.0.0.1 ./FahCore_a1.exe -dir work/ -suffix 01 -checkpoint 5 -verbose -lifeline 7979 -version 602'

[14:27:26]
[14:27:26] *------------------------------*
[14:27:26] Folding@Home Gromacs SMP Core
[14:27:26] Version 1.74 (November 27, 2006)
[14:27:26]
[14:27:26] Preparing to commence simulation
[14:27:26] - Ensuring status. Please wait.
[14:27:27] - Starting from initial work packet
[14:27:27]
[14:27:27] Project: 2605 (Run 8, Clone 393, Gen 39)
[14:27:27]
[14:27:27] Assembly optimizations on if available.
[14:27:27] Entering M.D.


And here it stops.

Processes executing :
[user@host smp]$ ps -ef |grep -v grep |grep -i fah
500 7979 7123 0 16:27 pts/0 00:00:00 ./fah6 -smp -verbosity 9
500 7984 7979 0 16:27 pts/0 00:00:00 ./mpiexec -np 4 -host 127.0.0.1 ./FahCore_a1.exe -dir work/ -suffix 01 -checkpoint 5 -verbose -lifeline 7979 -version 602
500 7986 7984 0 16:27 pts/0 00:00:00 ./FahCore_a1.exe -dir work/ -suffix 01 -checkpoint 5 -verbose -lifeline 7979 -version 602


//

[So_it_goes99]

No ideas ??

//