I fear there is something going on at your end, Ville.

The Special Application does have issues with some noise bombs when it comes to triplets & pulses & what goes where.

---

Grant
Darwin NT

Something's broken- Received-last-hour for both MB & AP have plummeted, as has the splitter output. All at the same time, around 21:20 UTC.
Although it looks like things might just be coming back to life again; splitter output is no longer 0, Returned-last-hour is climbing again.

I think that's usually the sign of everything being switched off and back on again, either to clear a blockage (some server getting stuck) or to install a new version of something.

No body hear me, but each time the total of Wu reaches 23 MM something weird happening. Can be just coincidence?

---

No body hear me, but each time the total of Wu reaches 23 MM something weird happening. Can be just coincidence?

Maybe. What are you totalling to get the 23M number?
The fact is, issues are occurring all the time, particularly with the Scheduler. And the backlog of Validation, Assimilation & Deletion and then Purging work doesn't help either.

You can see the Scheduler issues by looking at the In-progress numbers- people returning & reporting work, but not getting any replacement WUs.



Tues 22:30 UTC is the weekly outage, but after that there's blips at Thurs 02:00 & Thurs 18:00, then a bunch of big ones (several hours) Fri 01:00, another Fri 20:00 and the most recent Sat 10:00.

---

Grant
Darwin NT

No body hear me, but each time the total of Wu reaches 23 MM something weird happening. Can be just coincidence?

Maybe. What are you totalling to get the 23M number?

Everything? In-progress, WU & Results awaiting validation, assimilation, deletion & purging?
Could be that's the point where everything jams up, and then it takes a while for things to get going again once some of those numbers have reduced.

---

Grant
Darwin NT

No body hear me, but each time the total of Wu reaches 23 MM something weird happening. Can be just coincidence?

Maybe. What are you totalling to get the 23M number?

Everything? In-progress, WU & Results awaiting validation, assimilation, deletion & purging?
Could be that's the point where everything jams up, and then it takes a while for things to get going again once some of those numbers have reduced.

Yes everything, just sum all the lines and see. Could be just a coincidence but each time the total reaches 23 MM weird things starts to happening. When the totals down to 18-20 MM the system works normally. Please note i`m not saying that is a fact, was just an observation. As i say, could be just coincidence. Need more inside info to be sure.

---

No body hear me, but each time the total of Wu reaches 23 MM something weird happening. Can be just coincidence?

Maybe. What are you totalling to get the 23M number?

Everything? In-progress, WU & Results awaiting validation, assimilation, deletion & purging?
Could be that's the point where everything jams up, and then it takes a while for things to get going again once some of those numbers have reduced.

Yes everything, just sum all the lines and see. Could be just a coincidence but each time the total reaches 22-23 MM weird things happening. When the totals down to 18-20 MM the system works normally. Please note i`m not saying that is a fact, was just an observation.

My guess is that it's just the statistics dump that slows everything down while it runs.

---

Yes everything, just sum all the lines and see. Could be just a coincidence but each time the total reaches 22-23 MM weird things happening. When the totals down to 18-20 MM the system works normally. Please note i`m not saying that is a fact, was just an observation.

It would be good if it is the case. Sort out the Validation/ Assimilation/ Deletion/ Purge backlogs, and then they can increase the server side limits further & not have system issues (we can at least dream).

---

Grant
Darwin NT

My guess is that it's just the statistics dump that slows everything down while it runs.

Isn't that just 1once every 24hrs? The Scheduler issues can occur several times a day, and not always at the same times, and last for 30min to over 5 hours.

---

Grant
Darwin NT

My guess is that it's just the statistics dump that slows everything down while it runs.

Isn't that just 1once every 24hrs? The Scheduler issues can occur several times a day, and not always at the same times, and last for 30min to over 5 hours.

They've been running twice a day for some months. Approximately twelve hour intervals. Dump files go here, so it's easy to keep track.

---

Well, they're better than noise bombs (as long as they aren't noise bombs themselves), but it looks like there's a bunch of shorties going to hit the servers soon- 29ap11ah looks like it's pretty much all shorties.

---

Grant
Darwin NT

And the Scheduler's having another time out.
"Project has no tasks available" is the most common response, with a few 1 or 2 WUs handed out responses mixed in.

And website & forums getting laggy again.

---

Grant
Darwin NT

Well, the Scheduler is awake again, now it's the download servers having issues. Taking quite a while for the downloads to start, even after several retries.

---

Grant
Darwin NT

I hope the splitters get their act together soon, otherwise we'll be out of work in less than 15min.

---

Grant
Darwin NT

Uploads are slow and troublesome as well.

---

Uploads are slow and troublesome as well.

Been that way for ages, although it seems to be worse after each server issue when things are recovering, and even worse than usual after the last couple of server problems.
Mind you the return rate lately has been at record levels.

---

Grant
Darwin NT

It seems the Scheduler has checked out again. Can't get any New tasks, Webpages Slow, Time of last 'Results ready to send' results were 39 minutes ago, https://setiathome.berkeley.edu/show_server_status.php
Hopefully it will come back before my machines run outta work......AGAIN.

Things are looking better, but most machines are Still Not receiving tasks. This one is down to about a Dozen tasks left out of a Thousand, https://setiathome.berkeley.edu/results.php?hostid=6796479
It's about to be run out of work for the Third time in 24 hours.
Correction....it's now out of work....AGAIN.

I see we're presently recovering from yet another unscheduled Scheduler rest break.

---

Grant
Darwin NT

RTS = 39,968 this does not bode well for the recovery from the upcoming Tuesday outage.

---