I have recently had a spate of these errors on 3 different hosts and 4 different WU\'s

One Pop-up window contained:

Visual Fortran run-time error

forrtl: severe (38): error during write, unit 8, file
C:\\Data\\BOINC\\projects\\climateprediction.net\\hadsm3fub_jz96_005963821\\tmp\\pipe_dummy
Image PC Routine Line Source
hadsm3_um5.06_wi 00632852 Unknown Unknown Unknown
hadsm3_um5.06_wi 0062FAE4 Unknown Unknown Unknown
hadsm3_um5.06_wi 005EB622 Unknown Unknown Unknown
hadsm3_um5.06_wi 005EB23F Unknown Unknown Unknown
hadsm3_um5.06_wi 005D0C11 Unknown Unknown Unknown
hadsm3_um5.06_wi 00423B75 Unknown Unknown Unknown
hadsm3_um5.06_wi 0040A226 Unknown Unknown Unknown
kernel32 77E7EB69 Unknown Unknown Unknown

[ OK ]

All WU\'s showed Blue ice-worlds in graphics.

3 WU\'s either crashed or were aborted after several attempts to kick them into action.

The fourth WU I managed to get going again but while I was kicking it I noticed something very strange.


After the error message first popped up I clicked OK but it reappeared a few seconds later (all WU\'s did same).
I suspended tast and resumed - same result.
I suspended all tasks, exit/restart boinc - same result.

During the exit/restart of boinc I noticed that the windows tast manager was still showing the hadsm task (0% CPU)
After restarting boinc and resuming the WU the same hadsm task in task manager resumed action.

The graphics still showed a blue planet.

I though \"What the @#$%\" and just killed the task without suspending or exiting anything

BINGO - A new hadsm task started in task list (at 00:00 time), WU showed progress, no more error messages and when I checked the graphics everthing was looking normal (not blue).

This WU is still runing (now at 72%) and has sent 7 trickles (4 phase2 and 3 phase3) since I kicked it on sep 12.

Maybe others who experince problems should check for lingering tasks in task list?

Bruce