In the BOINC Manager, in Options > Computing preferences, there's a blue url at the top left for using web preferences.
In both of my computers, this points to the test site preferences, and I can't get it to change to the main site preferences.

Both sites have recently been upgraded to the latest, (I'm told), server version. I'm not sure if this has something to do with it.

My original and main problem, is that I can't get one of the computers to download any tasks from the main site. (May also be a problem with the other computer, and I just don't know it yet.)

The local prefs are correct, and I can't see anything else that would stop downloads.
But the event log says No tasks sent.

I've just uninstalled BOINC (Linux) and installed it again, with no change. Ubuntu is dishing up 7.6.31 if that's relevant.

Whatever is wrong is bound to be simple, and once I know, obvious.
But not at the moment.

In the BOINC Manager, in Options > Computing preferences, there's a blue url at the top left for using web preferences.
In both of my computers, this points to the test site preferences, and I can't get it to change to the main site preferences.

Both sites have recently been upgraded to the latest, (I'm told), server version. I'm not sure if this has something to do with it.

My original and main problem, is that I can't get one of the computers to download any tasks from the main site. (May also be a problem with the other computer, and I just don't know it yet.)

The local prefs are correct, and I can't see anything else that would stop downloads.
But the event log says No tasks sent.

I've just uninstalled BOINC (Linux) and installed it again, with no change. Ubuntu is dishing up 7.6.31 if that's relevant.

Whatever is wrong is bound to be simple, and once I know, obvious.
But not at the moment.

Ah, expected. Boinc uses the preferences of the last project (time stamp) web site that you changed them on. In your case that sounds like the test site. Update them on the main project web site and then issue an update on the main project from the manager and you should see that the site referenced in Boinc manager has changed to the main site.

That's what I thought, and I've done that several times to no avail.
But the "other" trick worked: let everyone know that you have a problem, and it disappears! :(

I switched the percentage option in a different venue, then changed to that venue and Updated.
And it worked.

That computer still won't download tasks though. It's as if the number of processors is set to less than one somewhere.
And I've just noticed that there are no messages about the number of processors changing.
If I change to a different venue with a different number of processors, it should say so.

This is what I get on the second computer:
Number of usable CPUs has changed from 8 to 5.

So there's still a second problem somewhere.

Please post the contents of the global_prefs.xml and global_prefs_override.xml files of the affected computer(s)

---

No override file, and the venue being used is "school"

************


<global_preferences>
<source_project>https://climateprediction.net/</source_project>
<source_scheduler>https://www.cpdn.org/cpdnboinc_cgi/cgi</source_scheduler>

<mod_time>1569463517</mod_time>
<max_ncpus_pct>100</max_ncpus_pct>
<cpu_usage_limit>100</cpu_usage_limit>
<run_on_batteries>0</run_on_batteries>
<run_if_user_active>1</run_if_user_active>
<run_gpu_if_user_active>1</run_gpu_if_user_active>
<idle_time_to_run>3</idle_time_to_run>
<suspend_if_no_recent_input>0</suspend_if_no_recent_input>
<suspend_cpu_usage>25</suspend_cpu_usage>
<work_buf_min_days>0.1</work_buf_min_days>
<work_buf_additional_days>0.5</work_buf_additional_days>
<cpu_scheduling_period_minutes>60</cpu_scheduling_period_minutes>
<disk_interval>60</disk_interval>
<disk_max_used_gb>30</disk_max_used_gb>
<disk_min_free_gb>0.1</disk_min_free_gb>
<disk_max_used_pct>100</disk_max_used_pct>
<ram_max_used_busy_pct>100</ram_max_used_busy_pct>
<ram_max_used_idle_pct>100</ram_max_used_idle_pct>
<leave_apps_in_memory>1</leave_apps_in_memory>
<vm_max_used_pct>75</vm_max_used_pct>
<max_bytes_sec_down>0</max_bytes_sec_down>
<max_bytes_sec_up>0</max_bytes_sec_up>
<daily_xfer_limit_mb>0</daily_xfer_limit_mb>
<daily_xfer_period_days>0</daily_xfer_period_days>
<dont_verify_images>1</dont_verify_images>
<confirm_before_connecting>0</confirm_before_connecting>
<hangup_if_dialed>1</hangup_if_dialed>
<venue name="home">
<max_ncpus_pct>12.5</max_ncpus_pct>
<cpu_usage_limit>100</cpu_usage_limit>
<run_on_batteries>0</run_on_batteries>
<run_if_user_active>1</run_if_user_active>
<run_gpu_if_user_active>1</run_gpu_if_user_active>
<idle_time_to_run>3</idle_time_to_run>
<suspend_if_no_recent_input>0</suspend_if_no_recent_input>
<suspend_cpu_usage>0</suspend_cpu_usage>
<work_buf_min_days>0</work_buf_min_days>
<work_buf_additional_days>0</work_buf_additional_days>
<cpu_scheduling_period_minutes>60</cpu_scheduling_period_minutes>
<disk_interval>60</disk_interval>
<disk_max_used_gb>50</disk_max_used_gb>
<disk_min_free_gb>0.1</disk_min_free_gb>
<disk_max_used_pct>100</disk_max_used_pct>
<ram_max_used_busy_pct>100</ram_max_used_busy_pct>
<ram_max_used_idle_pct>100</ram_max_used_idle_pct>
<leave_apps_in_memory>1</leave_apps_in_memory>
<vm_max_used_pct>75</vm_max_used_pct>
<max_bytes_sec_down>0</max_bytes_sec_down>
<max_bytes_sec_up>0</max_bytes_sec_up>
<daily_xfer_limit_mb>0</daily_xfer_limit_mb>
<daily_xfer_period_days>0</daily_xfer_period_days>
<dont_verify_images>1</dont_verify_images>
<confirm_before_connecting>0</confirm_before_connecting>
<hangup_if_dialed>1</hangup_if_dialed>
</venue>
<venue name="work">
<max_ncpus_pct>50</max_ncpus_pct>
<cpu_usage_limit>100</cpu_usage_limit>
<run_on_batteries>0</run_on_batteries>
<run_if_user_active>1</run_if_user_active>
<run_gpu_if_user_active>1</run_gpu_if_user_active>
<idle_time_to_run>3</idle_time_to_run>
<suspend_if_no_recent_input>0</suspend_if_no_recent_input>
<suspend_cpu_usage>0</suspend_cpu_usage>
<work_buf_min_days>1</work_buf_min_days>
<work_buf_additional_days>4</work_buf_additional_days>
<cpu_scheduling_period_minutes>60</cpu_scheduling_period_minutes>
<disk_interval>60</disk_interval>
<disk_max_used_gb>50</disk_max_used_gb>
<disk_min_free_gb>0.1</disk_min_free_gb>
<disk_max_used_pct>100</disk_max_used_pct>
<ram_max_used_busy_pct>100</ram_max_used_busy_pct>
<ram_max_used_idle_pct>100</ram_max_used_idle_pct>
<leave_apps_in_memory>1</leave_apps_in_memory>
<vm_max_used_pct>75</vm_max_used_pct>
<max_bytes_sec_down>0</max_bytes_sec_down>
<max_bytes_sec_up>0</max_bytes_sec_up>
<daily_xfer_limit_mb>0</daily_xfer_limit_mb>
<daily_xfer_period_days>0</daily_xfer_period_days>
<dont_verify_images>1</dont_verify_images>
<confirm_before_connecting>0</confirm_before_connecting>
<hangup_if_dialed>1</hangup_if_dialed>
</venue>
<venue name="school">
<max_ncpus_pct>100</max_ncpus_pct>
<cpu_usage_limit>100</cpu_usage_limit>
<run_on_batteries>0</run_on_batteries>
<run_if_user_active>1</run_if_user_active>
<run_gpu_if_user_active>1</run_gpu_if_user_active>
<idle_time_to_run>3</idle_time_to_run>
<suspend_if_no_recent_input>0</suspend_if_no_recent_input>
<suspend_cpu_usage>0</suspend_cpu_usage>
<work_buf_min_days>0</work_buf_min_days>
<work_buf_additional_days>0</work_buf_additional_days>
<cpu_scheduling_period_minutes>60</cpu_scheduling_period_minutes>
<disk_interval>60</disk_interval>
<disk_max_used_gb>50</disk_max_used_gb>
<disk_min_free_gb>0.1</disk_min_free_gb>
<disk_max_used_pct>100</disk_max_used_pct>
<ram_max_used_busy_pct>100</ram_max_used_busy_pct>
<ram_max_used_idle_pct>100</ram_max_used_idle_pct>
<leave_apps_in_memory>1</leave_apps_in_memory>
<vm_max_used_pct>75</vm_max_used_pct>
<max_bytes_sec_down>0</max_bytes_sec_down>
<max_bytes_sec_up>0</max_bytes_sec_up>
<daily_xfer_limit_mb>0</daily_xfer_limit_mb>
<daily_xfer_period_days>0</daily_xfer_period_days>
<dont_verify_images>1</dont_verify_images>
<confirm_before_connecting>0</confirm_before_connecting>
<hangup_if_dialed>1</hangup_if_dialed>
</venue>
</global_preferences>

But the event log says No tasks sent.

And while batch details says 500+ left to send out, clicking on unsent shows only 127 urls of tasks still to go out so If you don't get it sorted quickly you will have missed the boat for the current batch.

And while batch details says 500+ left to send out, clicking on unsent shows only 127 urls of tasks still to go out so If you don't get it sorted quickly you will have missed the boat for the current batch.

Actually quite a lot of batch (over 5,000) still to be uploaded so more bites of the cherry Les and anyone else chasing Linux work.

Whatever is wrong, I think it's associated with there being no mention of the number of CPUs in the Event log.
It may be an OS problem.
I'm going to look for a later version of the OS tomorrow. I think there should be one by now.
If that doesn't work, I'll dismantle the computer and use parts of it to build a new one.

It might be helpful to post the startup lines from the Event Log, where the CPU data would normally be registered following BOINC detection.

It would also be helpful to enable the <sched_op_debug> Event Log option before your next work fetch attempt, and post the output from that: without being too verbose, that does provide more information about what the client is requesting, and the server response.

Yes, I thought of the config stuff, and had a look. It's changed a lot since I looked last.
And I did put that <sched_op_debug> into the sample config file that was in among the BOINC stuff, and read it.
All that happened was that the event log then mentioned it, but there were no extra messages, which was disappointing.

At this point I gave up for a few hours, but then decided to look at my collection of OS disks.
I found one that was only a couple of versions earlier, so I spent the next few hours rebuild the OS and upgrading it to current. Which was the worst part of it.
Then I loaded up BOINC again, the easy part.
Next was to run it and see what happened.
Attached the "new" computer to the project, and it immediately used the default venue.
So I turned off getting new tasks and the Net access, set up the BOINC options the way that I wanted, went to the web site and changed the venue and then went to bed, because of the 1 hour backoff.

It was early in the morning before I woke up again, and eventually I decided to get up and run it again to see what happened.
No tasks left by this time, but at least I'd see if there was anything different in the event log.
Off went the work request, and back came the reply:

Scheduler request completed: got 8 new tasks.

And down came dozens of files.

That was a shock.
But it does seem to be working now, although it would have been nice to know just what was going on before.

But it does seem to be working now, although it would have been nice to know just what was going on before.

To do that, you would have needed to enable the logging and capture the data before destroying the previous installation.

I did, but there wasn't anything in the Event log.
I've used cc_config before at times, and there's always been lots of extra lines in the Event log, but not this time.
I must have done a reboot at one point as well, because I remember see the config stuff right at the top of the list.

Over the weeks I also removed and installed BOINC a couple of times without it helping, so it must have been something to do with the OS.