The work distribution issue has been resolved.


Message boards : News : The work distribution issue has been resolved.

Message board moderation

To post messages, you must log in.
AuthorMessage
Profile Kyong
Project administrator
Project developer
Project tester
Project scientist
Avatar

Send message
Joined: 9 Jun 12
Posts: 586
Credit: 52,667,790
RAC: 0
Message 9551 - Posted: 5 Jun 2026, 21:44:34 UTC
The work distribution issue has been resolved.

Due to the unusually high load during the BOINC Formula event, the server-side work feeder cache was being exhausted too quickly. This caused some clients to receive “no tasks available” messages even though work was available in the database.

The feeder and scheduler configuration has been adjusted, and the server is now sending work normally again.

Thank you for your patience and happy crunching!

Radim Vanco (FoxKyong)
ID: 9551 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile JohnDK

Send message
Joined: 18 Aug 23
Posts: 5
Credit: 4,095,488
RAC: 4,444
Message 9557 - Posted: 6 Jun 2026, 16:44:14 UTC
Sat Jun 6 18:40:27 2026 | Asteroids@home | Sending scheduler request: Requested by user.
Sat Jun 6 18:40:27 2026 | Asteroids@home | Reporting 4 completed tasks
Sat Jun 6 18:40:27 2026 | Asteroids@home | Requesting new tasks for CPU
Sat Jun 6 18:40:29 2026 | Asteroids@home | Scheduler request completed: got 0 new tasks
Sat Jun 6 18:40:29 2026 | Asteroids@home | Project has no tasks available

The problem with getting 0 new tasks even with plenty work available has been there long before the BOINC Formula event.
ID: 9557 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 1 Jan 14
Posts: 308
Credit: 37,021,958
RAC: 5,213
Message 9558 - Posted: 6 Jun 2026, 22:05:09 UTC - in response to Message 9551.  
The work distribution issue has been resolved.

Due to the unusually high load during the BOINC Formula event, the server-side work feeder cache was being exhausted too quickly. This caused some clients to receive “no tasks available” messages even though work was available in the database.

The feeder and scheduler configuration has been adjusted, and the server is now sending work normally again.

Thank you for your patience and happy crunching!

Radim Vanco (FoxKyong)


I hope you could automate the process so it doesn't run out again.
ID: 9558 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Keith Myers
Avatar

Send message
Joined: 16 Nov 22
Posts: 192
Credit: 226,490,136
RAC: 374,018
Message 9560 - Posted: 7 Jun 2026, 3:40:45 UTC - in response to Message 9557.  
Sat Jun 6 18:40:27 2026 | Asteroids@home | Sending scheduler request: Requested by user.

The problem with getting 0 new tasks even with plenty work available has been there long before the BOINC Formula event.

+100

A proud member of the OFA (Old Farts Association)
ID: 9560 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Michael H.W. Weber
Avatar

Send message
Joined: 23 Apr 13
Posts: 18
Credit: 12,919,628
RAC: 115,401
Message 9562 - Posted: 7 Jun 2026, 11:39:29 UTC - in response to Message 9551.  

Last modified: 7 Jun 2026, 11:40:12 UTC
The work distribution issue has been resolved.

Due to the unusually high load during the BOINC Formula event, the server-side work feeder cache was being exhausted too quickly. This caused some clients to receive “no tasks available” messages even though work was available in the database.

The feeder and scheduler configuration has been adjusted, and the server is now sending work normally again.

...unfortunately, nothing has improved.
The issue isn't resolved at all.

Michael.
President of Rechenkraft.net e.V.
ID: 9562 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
pascali

Send message
Joined: 3 Oct 25
Posts: 2
Credit: 176,585
RAC: 617
Message 9563 - Posted: 7 Jun 2026, 16:04:56 UTC
I still have this problem, is there anything I need to do to fix this.

Thanks
ID: 9563 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
pascali

Send message
Joined: 3 Oct 25
Posts: 2
Credit: 176,585
RAC: 617
Message 9564 - Posted: 7 Jun 2026, 16:13:29 UTC
Hooray, managed to get some new tasks!
ID: 9564 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
UBT - Timbo

Send message
Joined: 19 Jun 12
Posts: 13
Credit: 12,675,762
RAC: 26
Message 9565 - Posted: 7 Jun 2026, 16:25:42 UTC - in response to Message 9551.  
The work distribution issue has been resolved.

Due to the unusually high load during the BOINC Formula event, the server-side work feeder cache was being exhausted too quickly. This caused some clients to receive “no tasks available” messages even though work was available in the database.

The feeder and scheduler configuration has been adjusted, and the server is now sending work normally again.

Thank you for your patience and happy crunching!

Radim Vanco (FoxKyong)


Hi Radim

Thanks to yourself (and Josef?) for hopefully resolving this issue for EVERY Asteroids member, and not just for the Formula BOINC Sprint challenge.

It is important to everyone that the project runs without issues so that everyone can contribute to the project efficiently and without incurring solvable problems. :-)

regards
Tim
ID: 9565 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
WuJJ

Send message
Joined: 5 Aug 16
Posts: 5
Credit: 62,506,278
RAC: 99,339
Message 9566 - Posted: 7 Jun 2026, 18:18:05 UTC
I kinda wonder if there is some inefficiency on the BOINC server code. My smaller machines with lower buffer continued to fetch work through the entire period when my bigger machines that would report and request more task per request got timeouts.
ID: 9566 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
adict2jane

Send message
Joined: 19 Sep 25
Posts: 16
Credit: 102,985,818
RAC: 97,195
Message 9567 - Posted: 7 Jun 2026, 19:56:57 UTC - in response to Message 9566.  
I kinda wonder if there is some inefficiency on the BOINC server code. My smaller machines with lower buffer continued to fetch work through the entire period when my bigger machines that would report and request more task per request got timeouts.


I am seeing the same behavior. My AM5 9950x rigs are all dry and cannot connect/download new work. However, all of my other computers are able to connect and download work units reasonably well. Very odd in my opinion.
ID: 9567 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
tjmajka

Send message
Joined: 16 Nov 25
Posts: 1
Credit: 867,007
RAC: 80
Message 9591 - Posted: 21 Jun 2026, 23:22:10 UTC - in response to Message 9551.  
Ever since the distribution problem got resolved i have not gotten ANY WORK. I have reinstalled the latest version of BOINC which I was already running. Ideas?
ID: 9591 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Cowman

Send message
Joined: 23 Apr 26
Posts: 1
Credit: 261,582
RAC: 4,174
Message 9596 - Posted: 23 Jun 2026, 21:49:11 UTC
I haven't been able to get any new work recently myself.
ID: 9596 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote

Message boards : News : The work distribution issue has been resolved.