Well I have tried the trick a couple of cycles now on both Win7 systems who have been getting the no work is available messages for an hour or so. I clicked update after the first request had completed and clicked a couple more times after and all I get is the usual the system is not sending work because the update request is too recent.

Then on the next automatic request for work, it appears to decide to come down.

---

Grant
Darwin NT

I was about to hit the second machine and suddenly it straightened out by itself,

Sun Aug 6 23:19:49 2017 | SETI@home | Reporting 4 completed tasks
Sun Aug 6 23:19:49 2017 | SETI@home | [sched_op] CPU work request: 4713.88 seconds; 0.00 devices
Sun Aug 6 23:19:49 2017 | SETI@home | [sched_op] NVIDIA GPU work request: 158504.49 seconds; 0.00 devices
Sun Aug 6 23:19:54 2017 | SETI@home | Scheduler request completed: got 1 new tasks
Sun Aug 6 23:25:02 2017 | SETI@home | Reporting 5 completed tasks
Sun Aug 6 23:25:08 2017 | SETI@home | Scheduler request completed: got 0 new tasks
Sun Aug 6 23:25:08 2017 | SETI@home | Project has no tasks available
Sun Aug 6 23:30:16 2017 | SETI@home | Reporting 3 completed tasks
Sun Aug 6 23:30:22 2017 | SETI@home | Scheduler request completed: got 0 new tasks
Sun Aug 6 23:35:31 2017 | SETI@home | Reporting 3 completed tasks
Sun Aug 6 23:35:36 2017 | SETI@home | Scheduler request completed: got 0 new tasks
Sun Aug 6 23:35:36 2017 | SETI@home | Project has no tasks available
Sun Aug 6 23:40:44 2017 | SETI@home | Reporting 3 completed tasks
Sun Aug 6 23:40:49 2017 | SETI@home | Scheduler request completed: got 0 new tasks
Sun Aug 6 23:40:49 2017 | SETI@home | Project has no tasks available
Sun Aug 6 23:45:57 2017 | SETI@home | Reporting 3 completed tasks
Sun Aug 6 23:46:04 2017 | SETI@home | Scheduler request completed: got 0 new tasks
Sun Aug 6 23:46:04 2017 | SETI@home | Project has no tasks available
Sun Aug 6 23:51:16 2017 | SETI@home | Reporting 3 completed tasks
Sun Aug 6 23:51:17 2017 | SETI@home | Scheduler request completed: got 0 new tasks
Sun Aug 6 23:51:17 2017 | SETI@home | Project has no tasks available
Sun Aug 6 23:56:30 2017 | SETI@home | Reporting 4 completed tasks
Sun Aug 6 23:56:31 2017 | SETI@home | Scheduler request completed: got 0 new tasks
Sun Aug 6 23:56:31 2017 | SETI@home | Project has no tasks available
Mon Aug 7 00:01:44 2017 | SETI@home | Reporting 3 completed tasks
Mon Aug 7 00:01:46 2017 | SETI@home | Scheduler request completed: got 48 new tasks
Mon Aug 7 00:06:55 2017 | SETI@home | Reporting 4 completed tasks
Mon Aug 7 00:06:56 2017 | SETI@home | Scheduler request completed: got 5 new tasks
Mon Aug 7 00:12:09 2017 | SETI@home | Reporting 2 completed tasks
Mon Aug 7 00:12:10 2017 | SETI@home | Scheduler request completed: got 2 new tasks
Mon Aug 7 00:17:18 2017 | SETI@home | Reporting 3 completed tasks
Mon Aug 7 00:17:19 2017 | SETI@home | Scheduler request completed: got 3 new tasks
Mon Aug 7 00:22:33 2017 | SETI@home | Reporting 4 completed tasks
Mon Aug 7 00:22:34 2017 | SETI@home | Scheduler request completed: got 5 new tasks
Mon Aug 7 00:27:46 2017 | SETI@home | Reporting 3 completed tasks
Mon Aug 7 00:27:47 2017 | SETI@home | Scheduler request completed: got 2 new tasks

*shrugs*

I was about to hit the second machine and suddenly it straightened out by itself,
*shrugs*

Yeah.
It is rather borked.

---

Grant
Darwin NT

Man, I cannot get the linux system to get a full cache. The windows systems are fully up. I was down to 6 gpu tasks and managed to get a 22 task download after trying TBar's update technique. I have plateaued out at 150 gpu tasks out my 300 cache allotment. I have tried the flip-flip technique in Applications and multiple update requests during the normal scheduler request. I keep getting no tasks are available. This is not looking good for the outage tomorrow.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

I don't know why you have trouble getting tasks all the time, slow downloads?
I'm stuffed fuller than a Weight Watcher meeting at a buffet :D

BTW, I would like to welcome your computer "Bits and Pieces" to the Top 20 list!
Woo Hoo

My downloads are fine. It's just the wonky scheduler I have issues with.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

My downloads are fine. It's just the wonky scheduler I have issues with.

Yep,
Once we get them allocated, downloading is nice and quick.
The problem is just getting them allocated.

---

Grant
Darwin NT

And to add insult to injury ..... I seem to have trashed the entire gpu cache with 1 second computation errors. All have this in common.

Cuda error 'cufftPlan1d(&fft_analysis_plans[FftNum][0], FftLen, CUFFT_C2C, NumDataPoints / FftLen)' in file 'cuda/cudaAcc_fft.cu' in line 29 : invalid argument.

Restarted the system. Interesting that it had to re-download the master scheduling list. All the other files in the BOINC directory and the project directory all are present and look right. Something in BOINC went kablooey.

Starting to build the cache back up slowly.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

And to add insult to injury ..... I seem to have trashed the entire gpu cache with 1 second computation errors.

Ouch!
At least with all your Pendings alone it won't take long to build the cache back up.

---

Grant
Darwin NT

Yes, but my plan was to have enough tasks to make it through the outage. I'm not going to shepherd the system through the wee hours. Time to hit the hay. I guess the power consumption will be a bit less from 10AM onwards.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Yes, but my plan was to have enough tasks to make it through the outage. I'm not going to shepherd the system through the wee hours.

Given the number of Pendings, and the rate that system can churn out the work, you should be up against the server limits before the outage (as long as the Scheduler behaves of course).

---

Grant
Darwin NT

Sorry you didn't have time to load up Keith, but I see you have a full 400 tasks rith now on the Linux box.
The servers have been handing out nothing but Arecibo tasks to me (and no vlars) so it was a good time to load up :)
State: All (8506) · In progress (1064) · Validation pending (4275) · Validation inconclusive (158) · Valid (3004)

Locked and Loaded for maintenance.

Huh, just after 9 my time (7 Berkeley) and the site is still up? Weird. Not complaining mind you, but unusual to start maintenance this late, isn't it?

---

They were late last week too. Started right around this time I think ...

I'm waiting for them to figure out how to make it start late and end early..........LOL.
GPUs always running out of work during these epic outrages.

Meow.

---

"Time is simply the mechanism that keeps everything from happening all at once."

My 750Ti's are perfectly suited for 12h of maintenance. Just run the shorties through ahead of time, and they are good for ~13.5h

Well, the kitties pretty much keep the crunchers on autopilot around here, with them supervising of course.

---

"Time is simply the mechanism that keeps everything from happening all at once."

That was some outrageous outrage... about 23 hours. Of course now I am at work so can't micromanage. Darn.

---

Sorry everyone, this was entirely my fault. I had just gotten back from vacation, started the outage, and then managed to accidentally kill the outage. I discovered this much later and restarted it. My mind must have still been in the mountains.

---

Thanks for the info! And nice to see you around here... maybe the outage wasn't all bad then. :^)

---