7 errors over the last 3 days for this machine:
E000770_ 834C_ 003003413_ 0-- LEN0086 Error 6/28/09 20:27:19 6/29/09 13:13:23 10.83 125.1 / 0.0
E000770_ 674C_ 003004414_ 0-- LEN0086 Error 6/28/09 18:40:38 6/29/09 13:13:23 10.78 124.5 / 0.0
E000769_ 673C_ 00300830t_ 1-- LEN0086 Error 6/28/09 08:55:24 6/29/09 05:24:32 11.47 132.5 / 0.0
E000769_ 519C_ 003e0490v_ 1-- LEN0086 Error 6/28/09 07:22:45 6/29/09 05:24:32 11.20 129.4 / 0.0
E000768_ 562C_ 002x0420w_ 0-- LEN0086 Error 6/27/09 21:21:21 6/28/09 18:40:38 11.61 134.1 / 0.0
E000768_ 376C_ 005v03605_ 1-- LEN0086 Error 6/27/09 19:46:42 6/28/09 18:40:38 11.76 135.8 / 0.0
E000766_ 022C_ 002x0520h_ 0-- LEN0086 Error 6/26/09 22:37:34 6/27/09 19:46:42 11.17 129.0 / 0.0

The oldest 3 WUs have 3 errors on them already, with at least 1 more rep in progress.

OK, thanks for your quick reply.

My TCEP status so far for the last 3 days:
8 valid (2 crunching days),
20 PV (5.5 crunching days) and
52 failed (15 crunching days).

Just freeze it into Suspend mode till official word. If you're getting low on work, temporarily unfreeze so the client will go out to fetch more.

I have had a few of each 765/766/767 finish with errors like this:

<core_client_version>6.2.18</core_client_version>
<![CDATA[
<stderr_txt>
Calling initGraphics()
INFO: No state to restore. Start from the beginning.
called boinc_finish

</stderr_txt>
<message>
<file_xfer_error>
<file_name>E000766_755C_003e08503_0_0</file_name>
<error_code>-131</error_code>
</file_xfer_error>

</message>
]]>

I also have one 766 and one 768 which have successfully completed.

Well, if you want to set up a couple of 200+ hour Beta work units for me to crunch away at I wouldn't complain too much

Otherwise no, there shouldn't be a need to test anything (perhaps 1-2 units but that could probably easily be done in house).

I do hope though that something is being set up to give some credit to all the errored out work units. It would be really bad to have to give up many days (in some cases, weeks) of CPU time for something that probably wouldn't have errored out if things were running right on CEP's end. As the points/CPU time have no real world value (that I know of), it shouldn't matter if some extra ones were distributed for goodwill if nothing else.

Is there any estimation on the time frame needed to fix this problem? While I don't mind working on other projects (keeping my downloaded CEP files suspended & ready to go later), I do want to get back to work on the CEP (for a personal reason).

Thanks for helping sort out the chaos.