After several aborted CPDN runs, I have finally figured just what is causing the event. This last one happened with a file transfer error, which is on a RAID1 drive. It also happened around the time I was running defrag.

There is a secondary problem for me in that early aborts of a CPDN session due to errors rather than user action causes an annoying 24 hour wait for another work unit.

In one case, the download did not make the checksum match, and further downloads were delayed by 24 hours.

I am requesting two changes:
1. Detect a locked file and wait for acess rather than killing the run.
2. Detect download and run errors, and retry rather than simply abort.

BillN

---

After several aborted CPDN runs, I have finally figured just what is causing the event. This last one happened with a file transfer error, which is on a RAID1 drive. It also happened around the time I was running defrag.
...
I am requesting two changes:
1. Detect a locked file and wait for acess rather than killing the run.
...

That change would be nice. But in the meantime, it really is worth suspending the work (or even exiting BOINC altogether) before any kind of disk maintenance (defrag, scandisk, fsck, etc etc). I have always thought that defrag while working was a bit like trying to change the spark plugs on a running engine, it is amazing that it works at all, and not surprising when it blows up in your face.

---

I agree with River.

---

BOINC Wiki