FahMon (multi-platform app to monitor various F@h clients)

[toTOW]

uncle_fungus, it's time to update Fahmon to include this new psummary column : viewtopic.php?f=16&t=12204 ... and to include the "BigAdv" bonus formula in the PPD computation.

[SKeptical_Thinker]

Fahmon isn't tracking client properly. My setup is:

Code: Select all

Ubuntu 9.10 running in a VMware virtual machine
Host is Windows XP-32
Using VMware's player 3.0
Working directory accessed via samba

The symptom is that fahmon tracks the progress of the WU but not the FAHlog file. The status box stays yellow.

Here is what I see in the FAHlog:

Code: Select all

[14:56:32] Working on Unit 02 [November 18 14:56:32]
[14:56:32] + Working ...
[14:56:32] - Calling './mpiexec -np 4 -host 127.0.0.1 ./FahCore_a2.exe -dir work/ -suffix 02 -checkpoint 15 -verbose -lifeline 1785 -version 602'

[14:56:33] 
[14:56:33] *------------------------------*
[14:56:33] Folding@Home Gromacs SMP Core
[14:56:33] Version 2.10 (Sun Aug 30 03:43:28 CEST 2009)
[14:56:33] 
[14:56:33] Preparing to commence simulation
[14:56:33] - Ensuring status. Please wait.
[14:56:42] - Looking at optimizations...
[14:56:42] - Working with standard loops on this execution.
[14:56:42] - Files status OK
[14:56:53] - Expanded 4924168 -> 24360573 (decompressed 494.7 percent)
[14:57:00] Called DecompressByteArray: compressed_data_size=4924168 data_size=24360573, decompressed_data_size=24360573 diff=0
[14:57:01] - Digital signature verified
[14:57:01] 
[14:57:01] Project: 2662 (Run 2, Clone 170, Gen 51)
[14:57:01] 
[14:57:01] Entering M.D.
[14:57:07] Using Gromacs checkpoints

Here is what is actually in the log file:

Code: Select all

[12:49:39] + Working ...
[12:49:39] - Calling './mpiexec -np 4 -host 127.0.0.1 ./FahCore_a2.exe -dir
work/ -suffix 03 -checkpoint 15 -verbose -lifeline 6184 -version 602'

[12:49:39] 
[12:49:39] *------------------------------*
[12:49:39] Folding@Home Gromacs SMP Core
[12:49:39] Version 2.10 (Sun Aug 30 03:43:28 CEST 2009)
[12:49:39] 
[12:49:39] Preparing to commence simulation
[12:49:39] - Ensuring status. Please wait.
[12:49:40] Called DecompressByteArray: compressed_data_size=4835374
data_size=23979129, decompressed_data_size=23979129 diff=0
[12:49:41] - Digital signature verified
[12:49:41] 
[12:49:41] Project: 2669 (Run 15, Clone 74, Gen 110)
[12:49:41] 
[12:49:41] Assembly optimizations on if available.
[12:49:41] Entering M.D.
[12:49:47] Using Gromacs checkpoints
[12:49:51] 
[12:49:51] Entering M.D.
[12:49:57] Using Gromacs checkpoints
NNODES=4, MYRANK=0, HOSTNAME=mlm-desktop-HOME
NNODES=4, MYRANK=3, HOSTNAME=mlm-desktop-HOME
NNODES=4, MYRANK=2, HOSTNAME=mlm-desktop-HOME
NNODES=4, MYRANK=1, HOSTNAME=mlm-desktop-HOME
NODEID=2 argc=23
NODEID=3 argc=23
NODEID=0 argc=23
Reading file work/wudata_03.tpr, VERSION 3.3.99_development_20070618
(single precision)
NODEID=1 argc=23
Note: tpx file_version 48, software version 68

Reading checkpoint file work/wudata_03.cpt generated: Thu Nov 19 07:39:44
2009


NOTE: The tpr file used for this simulation is in an old format, for less
memory usage and possibly more performance create a new tpr file with an up
to date version of grompp

Making 1D domain decomposition 1 x 1 x 4
starting mdrun '22869 system'
27750000 steps,  55500.0 ps (continuing from step 27603090,  55206.2 ps).
[12:50:04] Resuming from checkpoint
[12:50:04] Verified work/wudata_03.log
[12:50:05] Verified work/wudata_03.trr
[12:50:05] Verified work/wudata_03.xtc
[12:50:05] Verified work/wudata_03.edr
[12:50:07] Completed 103090 out of 250000 steps  (41%)
[13:07:26] Completed 105000 out of 250000 steps  (42%)
[13:27:28] Completed 107500 out of 250000 steps  (43%)
[13:47:40] Completed 110000 out of 250000 steps  (44%)
[14:08:03] Completed 112500 out of 250000 steps  (45%)
[14:27:58] Completed 115000 out of 250000 steps  (46%)
[14:48:32] Completed 117500 out of 250000 steps  (47%)
[15:08:13] Completed 120000 out of 250000 steps  (48%)

I have uninstalled and reinstalled fahmon with and without retaining the existing data and clients.

Edit: I have noticed that it isn't updating the PPD, but it is updated the ETA and % complete.

Another edit: As of this morning the program is behaving correctly, again. I hate it when software gets sick, but at least it's better now.

[metal03326]

Does anyone know the default location of the report file created on crash? I didn't remembered it and I can't submit it to the site. I'm using XP.

[uncle_fungus]

It should be in your temp folder (I'm on Linux at the moment so I can't remember exactly where this is except that its somewhere under %APPDATA%)

[metal03326]

Found it - Local Settings\Temp. Thanks!

[bruce]

Does anyone know the default location of the report file created on crash? I didn't remembered it and I can't submit it to the site. I'm using XP.

The full report is nice, but you can get the essentials from the Event Viewer (in Admin Tools in Control Panel) even if the report is no longer available.

[metal03326]

I sent the full report, but I'll have in mind what you said when I loose my report. I doubt it'll be soon, after all I'm using FahMon for a long time and this is the first crash I had. 10x anyway.

[rickoic]

Perhaps this has been asked and answered before, but I haven't been able to find it, so point me to the post or please give me a clue.

I am running notfred's on different computers and trying to monitor them with fahmon. Trouble is that each notfred uses the same address \\fold-xxxx\c\etc\folding\1

Is there any way of changing the xxxx to a different address?

Tks
Rick

[toTOW]

Ask notfred, this is set up in his distribution ...

[Viper666]

Uncle any plan to add some bigadv bonus guessing to FAHmon?

[uncle fuzzy]

Perhaps this has been asked and answered before, but I haven't been able to find it, so point me to the post or please give me a clue.

I am running notfred's on different computers and trying to monitor them with fahmon. Trouble is that each notfred uses the same address \\fold-xxxx\c\etc\folding\1

Is there any way of changing the xxxx to a different address?

Tks
Rick

That's the format, but each of my notfred's has a different "xxxx". I've had 12 seperate VM installs, and all had a different xxxx. Unfortunately, FahMon doesn't always like that format, so some of mine are set to the IP for the client.


My current problem. I have one machine where I can't get it to work. This was my first folding machine, so it has worked fine for over 2 years and several upgrades.

XP Pro 32, Q6600, GTX260/216, 8800 GTS 512, 2 2-core VM notfred's (it also has a problem with a 4-core VM)

When I start it, it will display one or both GPUs, but the VMs always show red and Loading... using either address (\\fold- or IP). Once it opens, it never reloads, no matter what I do. I can close and reopen it, and it shows the current completion, but still won't update.

I've uninstalled and downloaded fresh copies several times, but it still does the same thing. I also have to kill fahmon.exe manually most times. Any ideas?

[DTHCND]

Aww I got it to run once but now when I type fahmon in a terminal fahmon instantly appears and stops responding... And back in the terminal it says...

Code: Select all

 (fahmon:26447): Gtk-WARNING **: gtk_widget_size_allocate(): attempt to allocate widget with width -5 and height 17

(fahmon:26447): Gtk-WARNING **: gtk_widget_size_allocate(): attempt to allocate widget with width -5 and height 17

then when I End it's task because it isn't responding it shows in the terminal

Code: Select all

Killed

HELP PLEASE!!!

Edit: Here are some pics of what is happening...

[Tynat]

Soo... no ideas on why mine is saving no window position, column and other GUI settings then I guess

Besides what I mentioned, no.

[DTHCND]

So no help?

[bruce]

attempt to allocate widget with width -5

I had a similar problem a long time ago (a much earlier version of FahMon). I never figured out how the window that FahMon had saved got corrupted but I was able to clear the corrupt data by deleting the config folder which put FahMon back to it's initial values and I was able to reconfigure it from there.

I see there are four files in that directory. You may be able to keep portions of your configuration (such as the pointers to you clients) by selecting which files to save and which ones to delete, but I don't know what is stored in each file so I can't help you with that.