Hey everyone,

Looks like the database has been down for a little over 30 hours and the boinc daemons are still struggling. Not entirely sure what happened, but I'm working through trying to fix it. I apologize for the slow progress being made, all of the major MilkyWay@home scientists, including myself are visiting collaborators in Canada and have been quite busy. I will try to get things fixed over night.

Jake

[edit]

Everything seems to be working now. I'll check back in the morning to make sure everything is still working.

[/edit]

I have successfuly reported all my finished tasks this morning and have received new CPU and GPU work.

Edit: GPU tasks are apparently still scarce, getting only few of them.

Tasks ready to send
295,536


9/24/2016 10:21:19 AM | Milkyway@Home | Sending scheduler request: Requested by user.
9/24/2016 10:21:19 AM | Milkyway@Home | Requesting new tasks for NVIDIA GPU
9/24/2016 10:21:21 AM | Milkyway@Home | Scheduler request completed: got 0 new tasks
9/24/2016 10:21:21 AM | Milkyway@Home | No work available


Anyone else having a problem.

24/09/2016 17:02:33 | Milkyway@Home | Sending scheduler request: Requested by user.
24/09/2016 17:02:33 | Milkyway@Home | Reporting 5 completed tasks
24/09/2016 17:02:33 | Milkyway@Home | Requesting new tasks for AMD/ATI GPU
24/09/2016 17:02:35 | Milkyway@Home | Scheduler request completed: got 0 new tasks
24/09/2016 17:02:35 | Milkyway@Home | Project has no tasks available

Thanks for looking.

Don't get any WU

24/09/2016 18:57:32 | Milkyway@Home | Sending scheduler request: To fetch work.
24/09/2016 18:57:32 | Milkyway@Home | Requesting new tasks for AMD/ATI GPU
24/09/2016 18:57:33 | Milkyway@Home | Scheduler request completed: got 0 new tasks
24/09/2016 18:57:33 | Milkyway@Home | No work available

---

Very similar. "No tasks available"

Yes, I don't get new tasks and send in work units stopt being validated after 13.30 utc.

Tasks ready to send 301,276

24/09/2016 18:23:24 | Milkyway@Home | Sending scheduler request: To fetch work.
24/09/2016 18:23:24 | Milkyway@Home | Requesting new tasks for CPU and Miner ASIC
24/09/2016 18:23:26 | Milkyway@Home | Scheduler request completed: got 0 new tasks
24/09/2016 18:23:26 | Milkyway@Home | No work available

Hey Everyone,

I've been busy working on the GPU compatibility issue. I thought this issue was mostly solved a month or so ago, but it seems to be popping back up again. The issue is the work generator refuses to make work even when there is a clear need for it. I plan on looking into the issue more after I push the new version of MilkyWay@home out on Monday.

I have an idea of how to fix this problem already so we will see how that works.

Sorry it took so long to respond to your posts. In the future if you want a quicker response from me, try replying on a relevant news post. I don't read number crunching daily.

Jake

Jake, looks like whatever you attempted to fix in the work generator is NOT working. Ready-to-send has plenty of work generated but nothing is going out when we request work. I got a full burst on one machine in the first two minutes that the project came back up, but nothing but dribbles since then on all three machines. Hope that you can fix the servers remotely from the conference.

---

"Jake, looks like whatever you attempted to fix in the work generator is NOT working. Ready-to-send has plenty of work generated but nothing is going out when we request work. I got a full burst on one machine in the first two minutes that the project came back up, but nothing but dribbles since then on all three machines. Hope that you can fix the servers remotely from the conference."

Same here. Server Status is showing a boat-load ready to send, but unable to fetch any of them.

EDIT - Just got a small batch of GPU units, but nothing for the CPUs

Hey everyone,

That's really strange. I'll give it a look when I get back to Troy tonight. Sorry it's taking so long to get responses.

Jake

Started getting CPU units late yesterday, but now the finished units wont upload and no new units of either kind are downloading. Here is a sample of my event log, from about the time this issue started..
9/25/2016 9:29:45 AM | Milkyway@Home | Starting task de_modfit_fast_15_3s_136_fixedangles3_2_1474693503_680879_0
9/25/2016 9:30:04 AM | Milkyway@Home | Sending scheduler request: To fetch work.
9/25/2016 9:30:04 AM | Milkyway@Home | Reporting 1 completed tasks
9/25/2016 9:30:04 AM | Milkyway@Home | Requesting new tasks for CPU and NVIDIA GPU
9/25/2016 9:31:06 AM | Milkyway@Home | Scheduler request failed: Error 408
9/25/2016 9:31:25 AM | Milkyway@Home | Message from task: 0
9/25/2016 9:31:25 AM | Milkyway@Home | Computation for task de_modfit_fast_15_3s_136_fixedangles3_1_1474693503_675771_0 finished
9/25/2016 9:31:25 AM | Milkyway@Home | Starting task de_modfit_fast_19_3s_136_ModfitConstraints1_2_1474693503_684691_0
9/25/2016 9:33:03 AM | Milkyway@Home | Sending scheduler request: To fetch work.
9/25/2016 9:33:03 AM | Milkyway@Home | Reporting 2 completed tasks
9/25/2016 9:33:03 AM | Milkyway@Home | Requesting new tasks for CPU and NVIDIA GPU
9/25/2016 9:33:20 AM | Milkyway@Home | Scheduler request failed: Error 408
9/25/2016 9:33:22 AM | Milkyway@Home | Message from task: 0
9/25/2016 9:33:22 AM | Milkyway@Home | Computation for task de_modfit_fast_15_3s_136_ModfitConstraints5_1_1474693503_676685_0 finished
9/25/2016 9:33:22 AM | Milkyway@Home | Starting task de_modfit_fast_19_3s_136_ModfitConstraints1_1_1474693503_684677_0
9/25/2016 9:33:58 AM | Milkyway@Home | Message from task: 0
9/25/2016 9:33:58 AM | Milkyway@Home | Computation for task de_modfit_fast_15_3s_136_ModfitConstraints5_4_1474693503_329076_1 finished
9/25/2016 9:33:58 AM | Milkyway@Home | Starting task de_modfit_fast_15_3s_136_ModfitConstraints5_2_1474693503_196240_3
9/25/2016 9:36:33 AM | Milkyway@Home | Sending scheduler request: To fetch work.
9/25/2016 9:36:33 AM | Milkyway@Home | Reporting 4 completed tasks
9/25/2016 9:36:33 AM | Milkyway@Home | Requesting new tasks for CPU and NVIDIA GPU
9/25/2016 9:36:43 AM | Milkyway@Home | Message from task: 0
9/25/2016 9:36:43 AM | Milkyway@Home | Computation for task de_modfit_fast_15_3s_136_fixedangles3_1_1474693503_421182_2 finished
9/25/2016 9:36:43 AM | Milkyway@Home | Starting task de_modfit_fast_15_3s_136_ModfitConstraints5_2_1474693503_687194_0
9/25/2016 9:37:52 AM | Milkyway@Home | Scheduler request failed: Error 408
9/25/2016 9:39:25 AM | Milkyway@Home | Message from task: 0

I am getting plenty of both CPU and GPU workunits now. Haven't done anything, they just started flowing on their own. Uploading and reporting finished workunits also work fine.

Yes, I finally have the full complement of tasks on all three machines myself. Don't know whether Jake fixed things remotely or the servers unplugged the stoppage on their own, whatever ...... I am happy again.

---

Some time gettings wus sometime not. If I try manual update boinc download new ones after a while I get no work and saw Milkyway@Home | No work available at log.

---

Hey everyone,

There is a huge backlog of Nbody work units so that might be causing some issues with it wanting to hand out separation work units. I am hoping the clear out soonish and that will solve most problems. In the mean time, I will be recompiling all of the BOINC and MilkyWay@home daemons on the server today using the latest BOINC versions. This will hopefully fix most of our GPU recognition problems. I'll make a new news post when that is done so I can get feedback.

Jake

Yes, there is a shortage of GPU workunits again. For the moment, I have plenty of N-body tasks for CPU, but nothing else.