Been getting frequent faults as follows (from Windows App Event Viewer). Started in BOINC 7.0.64 (x64) and persists in 7.2.28 (x64)

Most recent changes included replacing the video card with a faster one (radeon HD 7850) which involved uninstalling and reinstalling the supporting CIM sw AND installing the Steam (gaming) sw.

Would appreciate a fix or work around.

thank you

Faulting application name: hadcm3n_6.07_windows_intelx86.exe, version: 0.0.0.0, time stamp: 0x4d8c8033
Faulting module name: ntdll.dll, version: 6.1.7601.18247, time stamp: 0x521ea8e7
Exception code: 0xc0000005
Fault offset: 0x00037203
Faulting process id: 0x12f4
Faulting application start time: 0x01cedef3ecbea770
Faulting application path: E:\ProgramData\BOINC\projects\climateprediction.net\hadcm3n_6.07_windows_intelx86.exe
Faulting module path: C:\Windows\SysWOW64\ntdll.dll
Report Id: 2aaa20a5-4ae7-11e3-8150-20689d415a00

Hi Perini

Taking a few error listings from your various models:

Model crashed: ATM_DYN : INVALID THETA DETECTED.

This means that the model's planetary physics has failed.
(Atmosphere has suddenly disappeared; all/part of the planet has jumped to millions of degrees; or it's all turned into a ball of ice; etc)
You'd need to be a climatologist and understand what all the values for the various variables mean, to work out what happened.
It's one of the things that this project is about.

Model crashed: INITTIME: Atmosphere basis time mismatch

A problem with one of the many sets of data that are supplied to cpdn by the researchers, that's used to create a data set for downloading.
Nothing you can do about this.

WU download error: couldn't get input files:

This is from a regional model, and it means that it's an old model from way back, that a certain "BOINC feature" is taken it upon itself to re-issue.
This has been posted about in lots of places over the last few years.
Nothing you can do about it.

The drive cannot locate a specific area or track on the disk. (0x19) - exit code 25 (0x19)

Not sure about this. Perhaps another program got in the way of a HD access at a critical moment. Make sure that your anti virus program has been excluded from ALL scanning of BOINC and it's sub-folders.

Model crashed: INITDUMP: Wrong no of ocean prognostic fields

Another type of file mismatch from the researchers.

Reason: Out Of Memory (C++ Exception) (0xe06d7363) at address 0x7677C42D

Program problem.
I remember this from when we were beta testing, but don't remember the outcome.

- exit code -1073741205 (0xc000026b)

One of several Windows messages related to the graphics. Perhaps you started doing something which suddenly used a lot of memory at a critical moment, and the climate model failed.

*****************

On the machine that you haven't used for a while:

11:53:33 (3224): Can't acquire lockfile (32) - waiting 35s
11:53:45 (992): No heartbeat from core client for 30 sec - exiting

Possible some other program is using a hardware resource that the climate model needs.
Or your AV program has locked a file while it scans it.

WU download error: couldn't get input files:
<file_xfer_error>
<file_name>ocean_zbqy_1920_40_008255472_0.gz</file_name>
<error_code>-224</error_code>

This was back in March, when the server was having problems.
Nothing that you can do in these situations.

Model crashed: WRITDUMP: BAD BUFFOUT OF DATA tmp/pipe_dummy 2048
cpdnmonitor: error reading file C:\Documents and Settings\All Users\Application Data\BOINC/projects/climateprediction.net/hadcm3n_205j_1980_40_008255521/dataout/atmos_restart.day
cpdnmonitor: error reading file C:\Documents and Settings\All Users\Application Data\BOINC/projects/climateprediction.net/hadcm3n_205j_1980_40_008255521/dataout/ocean_restart.day
BUFFIN: C I/O Error ferror - Unit 41 - Return code = 32

Some other program stopping the file from being read?
Or perhaps the file was corrupted when it was being written.

There are a few more, but basically there's not much wrong with what you're doing.

Block the AV scanning as above, don't interrupt the hadcm3 models at their critical points, ALWAYS stop the models gently. i.e. if you're about to shut down the computer:
1) Set cpdn for No new tasks
2) Suspend each climate model, and wait until it says Suspended
3) Suspend BOINC
4) Exit from BOINC