Bonjour Lafason,
je n'ai aucune idée de la cause de ce problème auquel je suis, moi aussi, confronté épisodiquement.
En revanche la solution la plus efficace est:

• Si possible, rebooter la machine
  
• Contrôler ensuite dans Boinc manager si les WUs sont correctement reportées.
  
• Sinon, aller dans boinc manager sur la page des transferts en cours, et sélectionner une WU, cliquer sur "retry now"
  
• Attendre la fin du transfert et recommencer au point 3 jusqu'à l'envoi de toutes les WUs.

Bonne chance,
Yves

Hello Jean,
on my side, the situation was as following:

• Host on line with internet access (other hosts use the same network)
  
• Host is reachable using boinc manager remotely
  
• Without any particular reasons, boinc claimed that it is unavailable to contact WCG (see error messages below)
  
• The reporting queue grew with all completed WUs.
  
• boinc retry - unsuccessfully - transfer from time to time, days long (I was business traveling at this time).
  
• Manual retry of transfers did not succeed.
  
• After host reboot, for each completed WU, manual retry of transfer.
  
• Afterward, everything ran well again.

...
01-10-2009 00:20:37|World Community Grid|Started upload of CMD2_0068-RADIA.clustersOccur-2V5Y_A.clustersOccur_469_1272698_1272965_1272817_1272847_1_0
01-10-2009 00:20:59||Project communication failed: attempting access to reference site
01-10-2009 00:20:59|World Community Grid|Temporarily failed upload of CMD2_0068-RADIA.clustersOccur-2V5Y_A.clustersOccur_469_1272698_1272965_1272817_1272847_1_0: system connect
01-10-2009 00:20:59|World Community Grid|Backing off 3 hr 56 min 53 sec on upload of CMD2_0068-RADIA.clustersOccur-2V5Y_A.clustersOccur_469_1272698_1272965_1272817_1272847_1_0
01-10-2009 00:21:00||Access to reference site succeeded - project servers may be temporarily down.
...

More or less, within 3.5. years, each host (in two separate locations) did experience this problem at least once.
I am still using boinc 5.10.45 on W2K, WinXP (32 or 64).
In this particular case (W2K), the host was "fresh" rebooted around 5 days ago (otherwise, it could occur several weeks or months after the last reboot).
I don't have any idea about the root cause of this problem.
Cheers,
Yves

Hi Sekerob
I would also think something happened with DNS related components or service ! ... but ...

• Since the move of the computer center, last January, I maintain WCG IP-addresses locally in the host file.
  
• Several hosts use the same network without experiencing the same problem.
  
• Even if the problem would be caused by a router,
  
  • why should this problem "survive" several days long?
    
  • why did the problem persist after a reboot?
    
  • why only a manual action - "retry transfer" WU by WU - does help?

I repeat, I don't have any clue about what the reason is (but I am really interested to know it).
Have a nice Sunday,
Yves