climateprediction.net home page
hadsm3 4.02 crashing every so often.

hadsm3 4.02 crashing every so often.

Questions and Answers : Windows : hadsm3 4.02 crashing every so often.
Message board moderation

To post messages, you must log in.

AuthorMessage
old_user357

Send message
Joined: 7 Aug 04
Posts: 7
Credit: 14,787
RAC: 0
Message 264 - Posted: 7 Aug 2004, 2:08:42 UTC
Last modified: 7 Aug 2004, 2:30:56 UTC

It's happened twice so far, but the modeler is crashing every 0.07-8%. I exit out of BOINC to restart. This has happened twice so far. The specific WU in question is: 00en_000025512_0.

First crash @ 0.08% - restart @ 0.07%.
Second crash @ 0.14% - restart @ 0.13%.

XP SP1, Athlon 3000+, 512MB ram, gigs of free disk spac.e
ID: 264 · Report as offensive     Reply Quote
old_user357

Send message
Joined: 7 Aug 04
Posts: 7
Credit: 14,787
RAC: 0
Message 265 - Posted: 7 Aug 2004, 2:16:18 UTC
Last modified: 7 Aug 2004, 2:17:24 UTC

Crash @ 0.15%. appcompat.txt would have been included if the XML content could be displayed here. Restarting @ 0.15%
ID: 265 · Report as offensive     Reply Quote
Profile astroWX
Volunteer moderator

Send message
Joined: 5 Aug 04
Posts: 1496
Credit: 95,522,203
RAC: 0
Message 267 - Posted: 7 Aug 2004, 2:34:26 UTC

Hi, Agent Green,

Is your machine overclocked? This CPDN Model is VERY sensitive to overclocking. Models tend to crash early-on on unstable machines. (They may be stable for other applications, but....)

See UK_Nick's links for pointers to appropriate tests to run to determine stability, if overclocked.

Best of luck.
________________________________________________
We have met the enemy and he is us -- Pogo
ID: 267 · Report as offensive     Reply Quote
old_user357

Send message
Joined: 7 Aug 04
Posts: 7
Credit: 14,787
RAC: 0
Message 268 - Posted: 7 Aug 2004, 2:36:11 UTC - in response to Message 265.  

Crash @ 0.19% - restarting @ 0.19%
ID: 268 · Report as offensive     Reply Quote
old_user357

Send message
Joined: 7 Aug 04
Posts: 7
Credit: 14,787
RAC: 0
Message 269 - Posted: 7 Aug 2004, 2:38:23 UTC - in response to Message 267.  

> Is your machine overclocked? This CPDN Model is VERY sensitive to
> overclocking. Models tend to crash early-on on unstable machines. (They may
> be stable for other applications, but....)

Not in the least. I've never had any problems with any other intensive applications either (predictor / boinc / quake). I'll check the stability testing links, but I find it odd for it to crash so frequently.
ID: 269 · Report as offensive     Reply Quote
Pconfig

Send message
Joined: 5 Aug 04
Posts: 84
Credit: 76,646
RAC: 0
Message 274 - Posted: 7 Aug 2004, 4:44:35 UTC

I don't know if it's very hot in your place but how hot is your cpu?
ID: 274 · Report as offensive     Reply Quote
old_user357

Send message
Joined: 7 Aug 04
Posts: 7
Credit: 14,787
RAC: 0
Message 334 - Posted: 7 Aug 2004, 16:12:24 UTC - in response to Message 274.  

> I don't know if it's very hot in your place but how hot is your cpu?

The CPU holds steady around 60 deg C. I ran the memtest application over the entirety of last night running all tests for the express purpose of beating the snot out of the RAM, and there was nothing there. I'll run the Pi program suggested if it starts to go nutty when Phase II begins.
ID: 334 · Report as offensive     Reply Quote
Pconfig

Send message
Joined: 5 Aug 04
Posts: 84
Credit: 76,646
RAC: 0
Message 491 - Posted: 9 Aug 2004, 4:57:24 UTC - in response to Message 334.  

> The CPU holds steady around 60 deg C. I ran the memtest application over the
> entirety of last night running all tests for the express purpose of beating
> the snot out of the RAM, and there was nothing there. I'll run the Pi program
> suggested if it starts to go nutty when Phase II begins.
>
LOL, 60 degrees is very hot but should be stable. Strange, you ran some kind of cpu test?
ID: 491 · Report as offensive     Reply Quote
old_user50

Send message
Joined: 5 Aug 04
Posts: 3
Credit: 14,464
RAC: 0
Message 514 - Posted: 9 Aug 2004, 11:07:28 UTC - in response to Message 491.  

60 degrees is very hot, and CPDN crashes at lower temperatures than other things. Fortunately, this BOINC version doesnt crash the whole system (at least in my experience) like the original did. It may be wise for you to clean any dust out of your heat sink, see if you can get that temperature down a little
ID: 514 · Report as offensive     Reply Quote
old_user19867

Send message
Joined: 21 Sep 04
Posts: 2
Credit: 241,351
RAC: 0
Message 12463 - Posted: 10 May 2005, 10:55:07 UTC - in response to Message 265.  

Maybe this is another problem but i have experienced the workunit to crash aswell. I noticed it when my system was only running at 50% and cpdn was nowhere to be found in the processes even though BOINC said it was running.

For me i believe it has to do with viewing the graphic, because when i view it it doesn't come up. I just retried this but without luck. I accidentally stopped a cpdn process and it goes straight to reporting an error :S.

My system runs non-overclocked at 42C full load max 50C in summer. Stable Asus board and Intel processor and Kingston RAM. BOINC v2.25 btw.
ID: 12463 · Report as offensive     Reply Quote
Profile geophi
Volunteer moderator

Send message
Joined: 7 Aug 04
Posts: 2184
Credit: 64,822,615
RAC: 5,275
Message 12464 - Posted: 10 May 2005, 12:51:27 UTC - in response to Message 12463.  

> For me i believe it has to do with viewing the graphic, because when i view it
> it doesn't come up. I just retried this but without luck. I accidentally
> stopped a cpdn process and it goes straight to reporting an error :S.
>
> My system runs non-overclocked at 42C full load max 50C in summer. Stable Asus
> board and Intel processor and Kingston RAM. BOINC v2.25 btw.
>
Do you have an ATI card with recent drivers? If so, BOINC version 4.2x (or above?) will not correctly produce graphics displays. Some Intel graphics accelerators also have problems. Don't use the screensaver, and don't try to display the globe in the visualization. You can download the Advanced Visualization from the left hand column of links to view the model run. Sorry I don't have the link handy for the long thread in the Seti forums about this problem.
ID: 12464 · Report as offensive     Reply Quote
Arnaud

Send message
Joined: 3 Sep 04
Posts: 268
Credit: 256,045
RAC: 0
Message 12476 - Posted: 10 May 2005, 16:16:56 UTC

<a href="http://setiweb.ssl.berkeley.edu/forum_thread.php?id=12948">long thread in the Seti forums about this problem</a>
Arnaud
ID: 12476 · Report as offensive     Reply Quote
racinjimy

Send message
Joined: 19 Apr 05
Posts: 53
Credit: 6,325,436
RAC: 0
Message 12627 - Posted: 16 May 2005, 17:32:22 UTC - in response to Message 265.  

Astro:

Don't I know it, I thought I was stable (prime95 stable (12 hours) memtest and super pi stable) but BOINC corrupted my windows and BIOS.

I had some issues on my DFI NF4 with ram timings.

I am sorted now, overclocked and stable with water cooling :)
ID: 12627 · Report as offensive     Reply Quote

Questions and Answers : Windows : hadsm3 4.02 crashing every so often.

©2024 cpdn.org