I know it had happened before, don't know which project that was, but 
just now it happened again. A project changed it's server to another 
hoster, and thus the IP changed as well, this time it was 
radioactive@home. The WU crunched happily until it was finished, even an 
extra long one because of expected problems with a server move, but the 
connections with the project failed on the new server.

The WUs usually trickle every 20min or so, but that's nothing I would 
look out for, it doesn't give any red warning signs that trickles don't 
arrive, they only congest in the project folder. But as they didn't go 
home, I know it started immediately with the installation of the new 
server and the change of IP.

The crunched WU didn't finally upload as well, and a new one was not 
fetched, as well because the project server was "down" according to BOINC.

I asked for help in this thread here:
http://radioactiveathome.org/boinc/forum_thread.php?id=87&nowrap=true#807
and got as an advice to stop BOINC and restart it, so that the new IP 
would be somehow made known to BOINC.

I was a bit reluctant, as I had just 9h of checkpointless RNA on the 
clock, but as there was another 130h to go, I decided to ditch those 
crunch hours. After the restart, everything went fine. OK, I had to kick 
the WU and the project manually to upload and report, as some hours 
delay was already appointed, but it worked as it should.

So my question:
Why do I have to stop and restart BOINC to make it aware of the new IP, 
while my Firefox knew asap about it?

Greetz, Sänger
_______________________________________________
boinc_dev mailing list
[email protected]
http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev
To unsubscribe, visit the above URL and
(near bottom of page) enter your email address.

Reply via email to