At the moment there is over 1 million tasks available, but I am unable to get any, keeps saying

Requesting new tasks for CPU
Scheduler request completed: got 0 new tasks

Is there only Windows work available?

I am running Linux and I have reduced work load on the machines to idle cpu cores but still no cpu work downloads.

Thanks
Conan

The download works for all systems, linux, win, android, but not always,
for Android all calculation errors occur

---

Downloads may work but they are not granting me any work units. I have been waiting for over 2 hours on one machine and more than an hour on my other two and nothing is coming my way.
Hitting update does not do much.

I guess I will have to wait and see if anything comes my way. I have had a few come through earlier but would like some more please.

I still believe that more work units are being generated for Windows based computers than Linux based computers, maybe when enough have been generated for Windows then it will be the turn for Linux.

Conan

I see the same problem, Conan. But for only some of my machines. Some get work just fine, and some can get nothing. This includes windows machines too.

Also, I doubt any of the admins read this forum. They are all over on the discord server.

I have the same thing on my Windows and Linux PCs.

Hey all,

I monitor this forum as of recently (the last few days) and I'll do my best to keep up.

Part of the issue seems to be that our daily result quota was set to 500, which was fine for GPU work since most systems only had one GPU per system, but it's inadequate for CPU work given there could be a dozen or more cores all active.

It's set to 10000 for now, but I'm open to changing that if we still have problems.

We also had to increase our shmem size for the feeder to have a sufficient cache. It was previously not configured, so set to the default of 100.
With this many tasks, and running on CPU rather than GPU, the task to host ratio drastically changed compared to our previous projects.

Please let me know if we continue to have problems!

Thanks! All my machines have work now.

Hey all,

I monitor this forum as of recently (the last few days) and I'll do my best to keep up.

Part of the issue seems to be that our daily result quota was set to 500, which was fine for GPU work since most systems only had one GPU per system, but it's inadequate for CPU work given there could be a dozen or more cores all active.

It's set to 10000 for now, but I'm open to changing that if we still have problems.

We also had to increase our shmem size for the feeder to have a sufficient cache. It was previously not configured, so set to the default of 100.
With this many tasks, and running on CPU rather than GPU, the task to host ratio drastically changed compared to our previous projects.

Please let me know if we continue to have problems!

Hey all,

I monitor this forum as of recently (the last few days) and I'll do my best to keep up.

Part of the issue seems to be that our daily result quota was set to 500, which was fine for GPU work since most systems only had one GPU per system, but it's inadequate for CPU work given there could be a dozen or more cores all active.

It's set to 10000 for now, but I'm open to changing that if we still have problems.

We also had to increase our shmem size for the feeder to have a sufficient cache. It was previously not configured, so set to the default of 100.
With this many tasks, and running on CPU rather than GPU, the task to host ratio drastically changed compared to our previous projects.

Please let me know if we continue to have problems!

You can count on the improvement of the Android application, all tasks crash, Android 12/14

<core_client_version>7.24.1</core_client_version>
<![CDATA[
<message>
process exited with code 195 (0xc3, -61)</message>
<stderr_txt>
 wrapper (8.1.26018): starting
 wrapper: running ./java/bin/java (-jar pano121.jar --device 0 --start 25029152 --end 25029336)
 wrapper: created child process 19604
 ./java/bin/java exited; CPU time 0.015155
 app exit status: 0x6
 called boinc_finish(195)

</stderr_txt>
]]>

---

Yes, we just deployed a change (1.03) to give us more useful error information.

So please run at least one more task on your android device and that should hopefully give us what we need to fix it on android.


Thanks!

Yes, we just deployed a change (1.03) to give us more useful error information.

So please run at least one more task on your android device and that should hopefully give us what we need to fix it on android.


Thanks!

Android 14
1.03
https://minecraftathome.com/minecrafthome/result.php?resultid=9570081

<core_client_version>7.24.1</core_client_version>
<![CDATA[
<message>
process exited with code 195 (0xc3, -61)</message>
<stderr_txt>
 wrapper (8.1.26018): starting
 wrapper: running ./java/bin/java (-jar pano121.jar --device 0 --start 37822488 --end 37822672)
 wrapper: created child process 31742
OpenJDK 64-Bit Server VM warning: No monotonic clock was available - timed services may be adversely affected if the time-of-day clock changes
fdsan: attempted to close file descriptor 4, expected to be unowned, actually owned by FILE* 0x761802f1e0
 ./java/bin/java exited; CPU time 0.016491
 app exit status: 0x6
 called boinc_finish(195)

</stderr_txt>
]]>

---

How long is this job expected to run? A a few days? Weeks? Months?

Pano121 has as many tasks as are left in the queue right now. But if we don't get the correct value out of the results by the end, we plan to make some code changes and generate more work to try again soon.

HDWGH is done with this step after the last tasks finish, but there will be a future step that I'm told has an immense amount of crunching to do.
Vague estimates have it at ~1 million times the work, but that's just based on how many seeds there are to crunch and what we'd need to do with them, not a measure of how long it would actually take.

I would guess months for that next step, at minimum. And we may have to do dynamic work generation due to the scale.

We'll keep y'all posted on that as it gets closer to crunch time. :)