| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Discovering Dengue Drugs - Together
Thread: assigning work to not-always-connected users |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 4
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I run off-line. My WCG profile shows that I have specified about 38 hours as my 'interval between connects'.
Today my system was given several Discovering Dengue workunits (they are "retreads" - sent out again after some previous cruncher failed to report completion) having deadlines of about 33 hours. I hope the WCG project administrators realize that I __cannot__ guarantee that these workunits will be reported back in time by my system. If WCG does not wish to risk their handed-out work missing its deadlines, they ought not assign work having a short deadline to a target system whose specified 'interval between connects' __exceeds__ the time-until-deadline of that work. . |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello mikus,
Don't worry about it. Several people have commented over the holidays that the server has mistakenly assigned them to the 'reliable' list for fast turn-around. You'll soon be off it. Anyway, the only work units that need fast turn-around are Beta test units. Normally it is just a convenience to try to help people waiting for a quorum to complete. Lawrence |
||
|
|
MD-Crusher
Advanced Cruncher Joined: Jul 23, 2007 Post Count: 52 Status: Offline |
There seem to be 1 or more bugs in the deadlines computation of the "retreads" and to whom they are send. If you have actually been connecting more frequently, the server may have taken that as the gospel. I've been getting these 'BOINC needs to connect' popup messages and endless log entries every 60 seconds with connect failures. Would like to see these reconnect attempts increment.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
MD-Crusher, that behaviour is by design.
For local problems, BOINC keeps retrying. For server problems, BOINC backs off. Like it or not, this is how it is supposed to work. The messages could do with cleaning up, that's all (it's on the todo list). |
||
|
|
|