So I noticed the other day that an older machine of mine running 6.2.19 no longer gets work from S@H anymore. It makes scheduler requests, and gets told there is no work available. I was told to post over here and see if there are any ideas.

I turned on some debugging flags and here is a snippet from the message log:

2014-09-02 14:55:26 [work_fetch_debug] Request work fetch: Project backoff ended
2014-09-02 14:55:26 [work_fetch_debug] compute_work_requests(): start
2014-09-02 14:55:26 [work_fetch_debug] compute_work_requests(): cpu_shortfall 302400.000000, overall urgency Need immediately
2014-09-02 14:55:26 SETI@home [work_fetch_debug] best project so far
2014-09-02 14:55:26 SETI@home [work_fetch_debug] compute_work_requests(): work req 302400.000000, shortfall 301471.421109, urgency Need immediately
2014-09-02 14:55:27 SETI@home [sched_op_debug] Starting scheduler request
2014-09-02 14:55:27 [work_fetch_debug] time_until_work_done(): est 0.000000 ssr 100.000000 apr 0.996929 prs 100.000000
2014-09-02 14:55:27 SETI@home Sending scheduler request: To fetch work. Requesting 302400 seconds of work, reporting 0 completed tasks
2014-09-02 14:55:33 SETI@home Scheduler request succeeded: got 0 new tasks
2014-09-02 14:55:33 SETI@home [sched_ops_debug] Server version 705
2014-09-02 14:55:33 SETI@home Message from server: No tasks sent
2014-09-02 14:55:33 SETI@home Message from server: No tasks are available for SETI@home v7
2014-09-02 14:55:33 SETI@home Message from server: No tasks are available for AstroPulse v6
2014-09-02 14:55:33 SETI@home Message from server: Tasks for NVIDIA GPU are available, but your preferences are set to not accept them
2014-09-02 14:55:33 SETI@home Message from server: Tasks for AMD/ATI GPU are available, but your preferences are set to not accept them
2014-09-02 14:55:33 SETI@home Message from server: Tasks for Intel GPU are available, but your preferences are set to not accept them
2014-09-02 14:55:33 SETI@home Project requested delay of 303.000000 seconds
2014-09-02 14:55:33 SETI@home [sched_op_debug] Deferring communication for 1 min 29 sec
2014-09-02 14:55:33 SETI@home [sched_op_debug] Reason: no work from project
2014-09-02 14:55:33 SETI@home [sched_op_debug] Deferring communication for 5 min 3 sec
2014-09-02 14:55:33 SETI@home [sched_op_debug] Reason: requested by project
2014-09-02 14:55:33 [work_fetch_debug] Request work fetch: RPC complete
2014-09-02 14:55:33 [work_fetch_debug] compute_work_requests(): start
2014-09-02 14:55:33 [work_fetch_debug] compute_work_requests(): cpu_shortfall 302400.000000, overall urgency Need immediately
2014-09-02 14:55:33 SETI@home [work_fetch_debug] work fetch: project not contactable; skipping

Another user ran a test and saw the same thing as me (seen here) whereby 6.2.19 does not get work assigned for them, either.

The current theory is that in the recent server version change from 703 -> 705 (as emphasized above), something got broken for older clients, or there is now a minimum version requirement..?

Does anyone have any ideas or suggestions? I know upgrading the installed version will give me work once again, but then that wouldn't be fixing a potential problem, just applying a work-around. So.. is this an actual problem, or is this the intended result of the 705 upgrade?

When a server has a minimum version of BOINC setup, it will inform you of which version it requires at minimum to allow you to get work. Since it doesn't do that in this case, it's a thing that happened with the scheduler, probably a bug that crept in. At Seti, there is another person with the same version you have with the same problem.

---

And another user who contacted me privately/anonymously, with even more serious problems - which I have reported privately to the project team.

Apparently there were some fixes deployed with the scheduler on Thursday and this problem seems to be resolved now. The host in question has a full ~3.5-day cache once again.

Thanks for the help, and for the fix!