Reb,
was wondering earlier if this type of message received from within Boinc Manager by clients couldn't be made more accurate when, obviously, for a given Project/Sub Project (SRBase or else), the so said Project has a lot of WUs to distribute but doesn't for any good reason.
I've faced this message hundreeds of times w/o being able to understand this wasn't due to my own configuration (cache config, tasks in progress -total time- or whatever ...).
I don't super trust BM's scheduller to be honest but sometimes we get responses from project servers that aint that clear as well.
So basically, I'm asking if this is ez, at the project level, to warn the users in this type of cases that, ya, WUs are available but won't be sent at a given time (server side decision). This would prevent us from scratching our heads and, in some occasions, fuxoring our own configurations for no reason at then end
/ 
Reply
Quote
rebirtherVolunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 2 Jan 13
Posts: 7479
Credit: 43,686,081
RAC: 42,669
Reb,
was wondering earlier if this type of message received from within Boinc Manager by clients couldn't be made more accurate when, obviously, for a given Project/Sub Project (SRBase or else), the so said Project has a lot of WUs to distribute but doesn't for any good reason.
I've faced this message hundreeds of times w/o being able to understand this wasn't due to my own configuration (cache config, tasks in progress -total time- or whatever ...).
I don't super trust BM's scheduller to be honest but sometimes we get responses from project servers that aint that clear as well.
So basically, I'm asking if this is ez, at the project level, to warn the users in this type of cases that, ya, WUs are available but won't be sent at a given time (server side decision). This would prevent us from scratching our heads and, in some occasions, fuxoring our own configurations for no reason at then end
Yes, I noticed that before. There are 3 reasons for that:
- the workscript is taking too many connections
- you have reached the limit of tasks in progress
- the feeder has not enough WUs in cache to sending out due high work requests
If you have still enough WUs in cache then dont worry about it.
/ Reply
Quote
Hi Reb,
I have zero cache and get this message more often than WU's !
So maybe SRBase's cache is undersized or can't handle the traffic - maybe too many short WU's ?
Crossing my fingers that the situation can be improved :-)
/ Reply
Quote
rebirtherVolunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 2 Jan 13
Posts: 7479
Credit: 43,686,081
RAC: 42,669
Hi Reb,
I have zero cache and get this message more often than WU's !
So maybe SRBase's cache is undersized or can't handle the traffic - maybe too many short WU's ?
Crossing my fingers that the situation can be improved :-)
Yes, the short WUs are out very fast and the server cant generate enough cache for the feeder, Iam adding more WUs now so the situation could be better.
/ Reply
Quote