There is a ton of "noise bombs" going through currently. Likely the reason for the higher return rate.

---

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

A proud member of the OFA (Old Farts Association)

There is a ton of "noise bombs" going through currently. Likely the reason for the higher return rate.

I looked at my returned tasks just after you posted, and at that time about 25% were noisy.
Just checked again and now its about 50%.
Server stats are also up.
Results received in last hour ** 146,414 0m

I've been checking the stats periodically in between chores, and it hit a peak around 150k/ hour and appears to be slowly falling now.

There is a ton of "noise bombs" going through currently. Likely the reason for the higher return rate.

Yep.
Noticed a lot in my results, and still quite a few in my cache.

And of course the random Credits being allocated- Anything from 0.3 to 5 depending on wingmate.

---

Grant
Darwin NT

The results returned per hour number is on the rise again, guessing that means more noise bombs. Still in the 140k -150k range, so no panic.

Thanks to "noise bombs" we have completed over 1000 channels today. No need to panic

---

The results returned per hour number is on the rise again, guessing that means more noise bombs. Still in the 140k -150k range, so no panic.

Up to 184K185K187K188K now.

---

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

A proud member of the OFA (Old Farts Association)

Only a few noisy bombs, average time to complete tasks is 310s (5m:10s)

25/05/2019 08:00:07 | SETI@home | Sending scheduler request: To fetch work.
25/05/2019 08:00:07 | SETI@home | Reporting 1 completed tasks
25/05/2019 08:00:07 | SETI@home | Requesting new tasks for NVIDIA GPU
25/05/2019 08:00:09 | SETI@home | Scheduler request completed: got 1 new tasks

5m:10s later

25/05/2019 08:05:17 | SETI@home | Sending scheduler request: To fetch work.
25/05/2019 08:05:17 | SETI@home | Reporting 7 completed tasks
25/05/2019 08:05:17 | SETI@home | Requesting new tasks for NVIDIA GPU
25/05/2019 08:05:19 | SETI@home | Scheduler request completed: got 6 new tasks

5m:10s later

25/05/2019 08:10:27 | SETI@home | Sending scheduler request: To fetch work.
25/05/2019 08:10:27 | SETI@home | Reporting 7 completed tasks
25/05/2019 08:10:27 | SETI@home | Requesting new tasks for NVIDIA GPU
25/05/2019 08:10:29 | SETI@home | Scheduler request completed: got 7 new tasks

Up a bit more
Results received in last hour ** 185,875

The results returned per hour number is on the rise again, guessing that means more noise bombs. Still in the 140k -150k range, so no panic.

Up to 184K now.

And the WU awaiting Validation & Assimilation continue to climb, Ready-to-send continues to fall, and the splitter output continues to decline.
Looks like we've found the servers limits again.

---

Grant
Darwin NT

Ha ha. I have to scroll through 15 pages to find a page that isn't less than 50% noise bombs. Don't know whether they have had a receiver go bad at Green Bank or they just recorded when there was an extraneous noise source in the antenna's aperture.

---

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

A proud member of the OFA (Old Farts Association)

Just watched one of my GPUs go through 10 noise bombs in a row.

---

Grant
Darwin NT

Ha ha. I have to scroll through 15 pages to find a page that isn't less than 50% noise bombs. Don't know whether they have had a receiver go bad at Green Bank or they just recorded when there was an extraneous noise source in the antenna's aperture.

My 1st page of "Valid" has 17 noise bombs.

I think we may have found the "fail-over" point in the servers. All my hosts are not able to download tasks but go straight to backoffs.

---

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

A proud member of the OFA (Old Farts Association)

I think we may have found the "fail-over" point in the servers. All my hosts are not able to download tasks but go straight to backoffs.

not seen here. times BST

25/05/2019 08:42:15 | SETI@home | Sending scheduler request: To fetch work.
25/05/2019 08:42:15 | SETI@home | Reporting 1 completed tasks
25/05/2019 08:42:15 | SETI@home | Requesting new tasks for NVIDIA GPU
25/05/2019 08:42:17 | SETI@home | Scheduler request completed: got 1 new tasks
25/05/2019 08:42:19 | SETI@home | Started download of blc25_2bit_guppi_58340_62915_HIP43542_0100.25780.0.20.29.226.vlar
25/05/2019 08:42:22 | SETI@home | Finished download of blc25_2bit_guppi_58340_62915_HIP43542_0100.25780.0.20.29.226.vlar

I think we may have found the "fail-over" point in the servers. All my hosts are not able to download tasks but go straight to backoffs.

No such issues here, yet.

---

Grant
Darwin NT

A server request for one task is a lot more likely to go through vice the several hundred I ask for at each connection.

---

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

A proud member of the OFA (Old Farts Association)

A server request for one task is a lot more likely to go through vice the several hundred I ask for at each connection.

I've had my work reporting limited to 60 for ages now due to that issue.

---

Grant
Darwin NT

A server request for one task is a lot more likely to go through vice the several hundred I ask for at each connection.

I've had my work reporting limited to 60 for ages now due to that issue.

But even with reporting limited to 60, I can have 5 hosts times 60 = 300 tasks being requested. And that is if I wasn't down several hundred tasks per host already because of the download backoffs, so each host is going to ask for whatever the RTS buffer has in it each time to try and rebuild my cache.

I have my hosts reporting 100 tasks which works fine when the servers aren't under extreme duress like they are now.

---

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

A proud member of the OFA (Old Farts Association)

But even with reporting limited to 60, I can have 5 hosts times 60 = 300 tasks being requested.

?
The WU reported limit is a per host limit, the number of hosts reporting from a particular connection wouldn't make any difference (unless you're talking about a significant number of the total number of active hosts for the entire project).



Results received in last hour is almost 190k now.
Splitter output has picked up a bit, but still behind the present load, so Ready-to-send continues to plummet (just a bit slower than is was).
Edit-
And a few Validators & Assimilators are disabled, which wouldn't be helping clear the surge in returned work.

---

Grant
Darwin NT

I think we may have found the "fail-over" point in the servers. All my hosts are not able to download tasks but go straight to backoffs.

. . I have been assigned new work but the downloads are going into backoff ...

Stephen

:(