We're working hard on a new, much faster work generator: the project is growing and the current one won't keep up for long.

It ran into a problem last night. It has been stopped, work generation is back on the usual path, and the affected tasks were cancelled server-side — your clients will drop them on their own, nothing for you to do.

Impact. No validated results lost; the cancelled tasks had not started computing. Queues are refilling.

We're continuing to work on it to make it solid before putting it back in service.

In reply to benjamin's message of 2 Oct 2026:

... the cancelled tasks had not started computing.

Yes they had.

You're right, and I was wrong — sorry. I've checked the server records: of the tasks cancelled on 2 October, about 1,600 had already been computing. That's real work lost on your machines, and my earlier statement that they hadn't started was incorrect. The cause of the download failures is fixed: no download errors on anything sent since 10:36 CEST on 2 October. Thank you for calling it out.

I have over 100 tasks that have finished 100% but just sit there not uploading and take up space preventing new tasks from downloading. I'm guessing these are the bad tasks and I should just delete them

Did you get an error in event.log and check stderr.txt in slotfolder.

I deleted all my tasks on all my computers to start over. It looks like I'm still having the same issue. Tasks finish 100% then just do nothing. I'll look into that file to see if I can find an issue

2026-10-02 10:17:49 (15696): wrapper: running ../../projects/boinc.eternityathome.org_eternity/bw_runner_233.exe (--force-interleave --shard-index 258 --shard-count 4096 --force-jitter-start 2061351209 --force-jitters 2061353257 --threads 1 --node-cap 2000000000 --seconds 0 --exact-endgame 244:50000000 --endgame-probe 4 --endgame-ladder 4:236:1000000000,3:230:6000000000,2:224:64000000000,1:216:80000000000 --ring-breaks 236 --sigma-tail --no-sigma-main --ncon2-direct 1 --mrv-dem-inc 1 --main-fast 1 --min-save 463 --breaks 192,197,201,206,211,216,221,225,229,233,237,241,246,250 --root-allow exploration_allow.txt --seed 6152026 --epoch cloudflare-boinc-explore --tourney tour)
2026-10-02 10:17:49 (15696): wrapper: created child process 3212
bw_runner: progress 0/80 (0%), restored runtime 0 s
bw_runner: progress 3/80 (3%), restored runtime 0 s
bw_runner: progress 7/80 (8%), restored runtime 115 s
bw_runner: progress 10/80 (12%), restored runtime 164 s
bw_runner: progress 14/80 (17%), restored runtime 222 s
bw_runner: progress 18/80 (22%), restored runtime 275 s
bw_runner: progress 22/80 (27%), restored runtime 335 s
bw_runner: progress 26/80 (32%), restored runtime 390 s
bw_runner: progress 30/80 (37%), restored runtime 445 s
bw_runner: progress 33/80 (41%), restored runtime 501 s
bw_runner: progress 37/80 (46%), restored runtime 558 s
bw_runner: progress 41/80 (51%), restored runtime 603 s
bw_runner: progress 45/80 (56%), restored runtime 713 s
bw_runner: progress 49/80 (61%), restored runtime 770 s
bw_runner: progress 53/80 (66%), restored runtime 827 s
bw_runner: progress 57/80 (71%), restored runtime 885 s
bw_runner: progress 61/80 (76%), restored runtime 939 s
bw_runner: progress 65/80 (81%), restored runtime 996 s
bw_runner: progress 69/80 (86%), restored runtime 1054 s
bw_runner: progress 73/80 (91%), restored runtime 1113 s
bw_runner: progress 77/80 (96%), restored runtime 1170 s
bw_runner: progress 80/80 (100%), restored runtime 1276 s
bw_runner: progress 80/80 (100%), restored runtime 1337 s
bw_runner: progress 80/80 (100%), restored runtime 1391 s
bw_runner: progress 80/80 (100%), restored runtime 1455 s
bw_runner: progress 80/80 (100%), restored runtime 1506 s
bw_runner: progress 80/80 (100%), restored runtime 1564 s
bw_runner: progress 80/80 (100%), restored runtime 1620 s
bw_runner: progress 80/80 (100%), restored runtime 1681 s
bw_runner: progress 80/80 (100%), restored runtime 1786 s
bw_runner: progress 80/80 (100%), restored runtime 1846 s
bw_runner: progress 80/80 (100%), restored runtime 1904 s
bw_runner: progress 80/80 (100%), restored runtime 1963 s
bw_runner: progress 80/80 (100%), restored runtime 2019 s
bw_runner: progress 80/80 (100%), restored runtime 2075 s
bw_runner: progress 80/80 (100%), restored runtime 2134 s
bw_runner: progress 80/80 (100%), restored runtime 2248 s
bw_runner: progress 80/80 (100%), restored runtime 2306 s
bw_runner: progress 80/80 (100%), restored runtime 2368 s
bw_runner: progress 80/80 (100%), restored runtime 2413 s

This is turning into a nightmare for me. I'm going to switch all my computers to another project if this isn't fixed soon

Was something done?
Tasks that were stalled seem to be uploading and reporting now

Did this run on win or linux?

Guess it wasn't fixed. The computer I was sitting at did upload the tasks but other computers are still stuck.
Example, I just checked a 9950X system and all 32 tasks are finished but new tasks won't start because the finished tasks won't upload/report

That file was from a Win11 computer with an Intel 13900K cpu
The stuck 9950X system is also running Win11

Can you check the windows event.log if there is something?

In reply to rebirther's message of 2 Oct 2026:

Can you check the windows event.log if there is something?

I looked at the event log but don't see anything. I never go into that log normally, so I don't really know what to look for.
All my Windows computers are set up the same and only use Windows Defender.

vThanks for the log, ChelseaOilman. It shows exactly what's going on. Nothing is stuck and no work is lost. The Windows app shows progress wrongly: it counts the search roots in a task, but each task has about two tickets per root. So the bar hits 100% roughly halfway through, and the task keeps computing for about the same time again before it finishes and uploads. Your log confirms it: 80/80 at about 21 minutes, and the solver was still writing results after that. The tasks that "suddenly uploaded" just finished normally. Please don't abort them, or the work done so far is lost. I'm fixing the progress display now and will post here when the update is out. Sorry for the confusion on top of last night's problems.

In reply to benjamin's message of 2 Oct 2026:

vThanks for the log, ChelseaOilman. It shows exactly what's going on. Nothing is stuck and no work is lost. The Windows app shows progress wrongly: it counts the search roots in a task, but each task has about two tickets per root. So the bar hits 100% roughly halfway through, and the task keeps computing for about the same time again before it finishes and uploads. Your log confirms it: 80/80 at about 21 minutes, and the solver was still writing results after that. The tasks that "suddenly uploaded" just finished normally. Please don't abort them, or the work done so far is lost. I'm fixing the progress display now and will post here when the update is out. Sorry for the confusion on top of last night's problems.

Ok, thanks for looking into it
Update: The 9950X system just uploaded tasks and it's working on new tasks

Good to hear the 9950X is back to normal. Quick update: the progress bar is fixed for every task sent since this evening, so new tasks now show their real progress. Tasks downloaded before that may still hit 100% early, but they finish and upload normally. Thanks again for the log, it made the cause easy to find.