I hope we are not going to start repeating the unscheduled Sunday outage.

I wonder whether reverting the server code to resolve the pre-Christmas problems, might have also re-instated the September/October Sunday outages? If so, that's good news - because we know that Eric has already found a solution for that.

I hope that is the case. Because it should be a quick fix since they have dealt with it before. But only time will tell and we have to wait till the Seti team is awake, coffee filled and ready to deal with this issue.

---

Is anyone getting new work? I am not.

Miklos

---

Getting worse, no uploads now.

---

Mike Bader
BOINC V7.16.5
http://setiathome.berkeley.edu/team_join_form.php?id=5 - Join Our International Team
[img]http://boinc.mundayweb.com/one/stats.php?

Got a note back from Eric. Seems like he agrees it's likely to be related to the pre-Christmas server reversion, and he thinks he's got it narrowed down. No promises of a restart, but he doesn't sound too worried.

bumping this to the top of the list so people will read first, and stop creating new threads. The server is down at the moment. Nothing to do about it. try some other projects, or save electricity, your choice.

edit: Thanks for the update Richard !! See post below.

And we're back up and running! Reported completed work, got new work in return.

Good work.

---

I just reported 107 tasks and got 101 in return.

---

Good work.

+1

---

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

A proud member of the OFA (Old Farts Association)

. . Yep, another weekend spoiled for Eric or whoever gave the kick, but why was Murphy so damned right ...

. . My problem here now is that while I am being given new work all the downloads are stalling :(

Stephen

:(

{Edit} - OK, the phantom of the fora strikes again, I wrote about the stalled downloads (which had reached into the hundreds) and now they are flowing as they should (more or less).

. . NOPE! Still happening on the other machines ...

<shrug>

Hi All

Today Seti has stacked up completed work units waiting for the server to accept them. I had about 120 of them waiting. They vanished without showing that they were uploading. Will I have been credited with the work?

But now the server is out of tasks to send...oh well, at least something is happening.

---

Caches are all full again here.

Cheers.

{Edit} - OK, the phantom of the fora strikes again, I wrote about the stalled downloads (which had reached into the hundreds) and now they are flowing as they should (more or less).

. . NOPE! Still happening on the other machines ...

<shrug>

Thanks for posting about the stalled downloads.....My stalled ones finally finished :)

---

{Edit} - OK, the phantom of the fora strikes again, I wrote about the stalled downloads (which had reached into the hundreds) and now they are flowing as they should (more or less).

Thanks for posting about the stalled downloads.....My stalled ones finally finished :)

. . Glad to be of service ... 8^}

Stephen

. .

Data Distribution State	SETI@home v7 #	Astropulse #	SETI@home v8 #	As of*
Results ready to send	0	0	98	0m
Current result creation rate **	0/sec	0.0938/sec	7.0673/sec	5m

Do we have a problem?

<edit> Forget about the Current Creation Rate rises to 77/sec

---

Yes. SETI had a problem today, but it's now resolved. It'll take a little while for the resulting mess - congestion - to clear up, but it'll settle down in a day or two.

When a task completes, two things happen. First, a file containing the result data is uploaded. That was going on normally all day, with one file uploading for each task as soon as it was complete - you probably didn't even notice them.

Secondly, BOINC 'reports' the task - does the housekeeping to say how long it took, etc. These reports can be stacked up and transferred in bulk - in fact, it's more efficient to do it that way. The report can't be done unless the upload has already taken place. So if they've gone, you know that everything is complete.

Results ready to send 0 0 1,595
This is not good

---

Results ready to send 0 0 1,595
This is not good

Current result creation rate ** 0.0613/sec 60.2323/sec

To keep the caches filled we need something > 40/sec. So >60/sec is fine, the splitters are just working hard to fill the new cache limits of 1000`s of hungry hosts..

Unless something else changes all will be returning to normal in few hours.

---

Results ready to send 0 0 1,595
This is not good

Current result creation rate ** 0.0613/sec 60.2323/sec

To keep the caches filled we need something > 40/sec. So >60/sec is fine, the splitters are just working hard to fill the new cache limits of 1000`s of hungry hosts..

The problem is that it needs to sustain that output, and in reality, it can't. You might get bursts over 50, but the average output is often a lot less.
Ready to send is now down to 1,139.

Edit- big jump to 5,290.
And big jumps are usually the result of no work going out. Hopefully it was just a case of a big gap between updates.

---

Grant
Darwin NT