Unless you did a clean install the messages can be recovered from various files. Discussion is covered in the Start Here forum, BOINC: Message & Error Log Reporting topic.

I don't understand why I'm having problems with this project. Three more work units that don't look very promising.........


Anyone have any ideas?


QX6700
3GB RAM
208GB hard drive space (installed on seperate drive, not on same drive as OS).

The QX6700 is a 4 core CPU. It has 8MB of L2 cache.

I'll try suspending 2 and see what happens. I'm willing to try almost anything. I didn't chose to get 3 in a row. That's just the way it happened.


Thanks for the advice. If you have any other ideas let me know.

If you don't mind walking me through how to find the information you want, I'll get it for you.

checkpoint interval and page fault delta - I don't know what they are or where to find them.

By my own counts of the logs for 3 at the default 60 second minimum, 50 or so are written per job (See Matrix post in Start Here forum for the counts of all WCG projects). That makes them about 14 minutes apart on my Q6600, 40 minutes on the P4, but the latter produced 6.5 billion PFs on a single non stop run, with some delta. I've had only a 50% validation rate on the 6 i had in a series on the P4 (1.3gb ram allowed), one sitting still in inconclusive and looking in the individual logs on the Result Status page, 2 had heartbeat issue, 3 had restarts/resumes, but more concerning, 1 had a smooth single run and went invalid not withstanding. Not very good. The 1 Q6600 job ran in 7 hours on Vista and validated. Clean single run.


Update: The remaining inconclusive validated so have an exact 50% 'invalid' rate, but all completed on the P4 in 17 to 21 hours and compared to all in quorum had 2 to 3 times longer run times!