Thanks Sekerob,

From your lead I have looked at slots and learned a little more.
I note that in Slots folder, I see 40 objects in 0 folder and nothing at all in 1 folder. Is that proper?

Done a reset. Sorry about the aborted WU's Mr Server.
Then a repair install and reboot.
Grabbed a CEP for each core. One has exited 29 after .08hrs, the other is at 35 minutes and running as I post.

I can now confirm I am user/all users install.

Skt A machine RAM is currently running 57% load on Rice.

Cheers

Hello Didactylos,

You could well be right that it is version related.

However I arrived at 6.6 quite innocently. I started out Boinc crunching at Einstein, until they enlarged their WU's beyond my capacity. So I went looking for something else to do and found WCG. As I already had Boinc I didn't need to look at the WCG version, and don't recall being aware of its existence at the time.
My earlier practice of updating the client from time to time continued, until I reached 6.6.20 at which time I learned here in the forums of the preference WCG has for one version. I know I could do a roll back on all machines, but there are things I like about 6.6.20 over earlier versions.

I guess I'll just solve or tolerate any problems until WCG catches up with me.

I, too, had a similar experience a couple days ago. Lost (errored) 4 pages of cep wus in a matter of about an hour. I was using 6.2.28 on a clean load of xp sp2.
I had been using this machine for cep and other projects before with no problems and changed the version of boinc to wcg 6.2.28 because of forum recommendation.
Decided later to go strictly cep to get my emerald badge and then that is when the problem struck. I did get four wus to about 4 hrs (uncompleted) then all hell broke loose.
I have only avg running and basic copy of firefox 3.5. This is a wcg machine.
I did have the boinc directories on the D: instead of the C:.
I did a total removal and installed boinc 6.47 and have had no problems after four wus and working on the next four now.
This is probably a moot point at this time being that cep on the downhill slide, but I find this interesting. And a little frustrating under these conditions.



Result Log

Result Name: E000866_ 500C_ 009f02001_ 0--
<core_client_version>6.2.28</core_client_version>
<![CDATA[
<message>
The system cannot write to the specified device. (0x1d) - exit code 29 (0x1d)
</message>
<stderr_txt>
Calling initGraphics()
INFO: No state to restore. Start from the beginning.
Encountered error. Exiting.

</stderr_txt>
]]>

Hello I'm also using 6.6.36 (windows 7, core i7) right now running 8 concurrent CEP with no problems.

Always running BOINC in advanced view, never using the show graphics and not using the screensaver.

In my experience CEP is the project most sensitive to hardware not 100% stable. I had alot of other WCG projects running fine but CEP errored out. A little more CPU corevolt solved that problem.

I'm using prime95 and IntelBurnTest to check for stability problems.

When I have had problems with CEP it's been broken batches or unstable hardware.

Not saying there never was/will be a version problem, just sharing my experience with CEP.

Apparently one of the tasks didn't like the re-boot after installing Microsofts latest patches, and nearly immediately errored-out:

29.07.2009 15:13:30|World Community Grid|[cpu_sched] Starting E000882_714C_009k04412_0(resume)
29.07.2009 15:13:30|World Community Grid|Restarting task E000882_714C_009k04412_0 using cep1 version 632
29.07.2009 15:13:30|World Community Grid|[cpu_sched] Starting E000884_415C_009l0650m_0(resume)
29.07.2009 15:13:30|World Community Grid|Restarting task E000884_415C_009l0650m_0 using cep1 version 632
29.07.2009 15:15:40|World Community Grid|[sched_op_debug] Deferring communication for 1 min 0 sec
29.07.2009 15:15:40|World Community Grid|[sched_op_debug] Reason: Unrecoverable error for result E000882_714C_009k04412_0 (The system cannot write to the specified device. (0x1d) - exit code 29 (0x1d))
29.07.2009 15:15:40|World Community Grid|Computation for task E000882_714C_009k04412_0 finished

The other result resuming at same time haven't been finished yet, but seems to be running normal, and has been doing this for over 3 hours now.

Oh, and the error-log is:

<core_client_version>5.10.45</core_client_version>
<![CDATA[
<message>
The system cannot write to the specified device. (0x1d) - exit code 29 (0x1d)
</message>
<stderr_txt>
Calling initGraphics()
INFO: No state to restore. Start from the beginning.
Calling initGraphics()
Encountered error. Exiting.

</stderr_txt>
]]>