Just had one of those rare, random events where host 2901600 got through to the scheduler.

It was (if I've counted them right) allocated 69 new tasks - more than two-thirds of a feeder cache. Mostly shorties, but heck - who can be choosy these days?

Saw the download block on BoincView, and got downstairs quickly enough to watch the downloads in real time. Absolutely smooth, fast (over 100 KB/sec) and with minimal pauses between files. Just one failed connection in the whole batch.

To my mind, that takes the focus off the router, Bay area networking, SYN floods and so on: unless it's a very sophisticated malware attack which is only targeting two of the four public-facing servers on the Hurricane Electric line (or possibly three - all my downloads came from Bane on 208.68.240.18 - I have no evidence of Vader's state of health from this single event).

Separarately, I dug out a copy of Wireshark this morning, on the same machine - which has just a few uploads to clear. All the SYN, SYN+ACK stuff seemed to clear through promptly and consistently, but all the attempts I watched eventually - at different stages in the process, and after varying delays - received RST+ACK from Bruno, and everything stopped. Ned?

Well I'm on empty and I still can't upload or report, So I'm stuck and do another project? No, Not while I have undelivered data, As My habit is to detach from one project and then attach to another, In any case I can't resolve seti according to ping.

---

The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's

Hmmmm.... uploading and reporting are still malfunctioning. It seems to me that someone at SETI would notice that "Results received in last hour" are consistently below normal etc, etc, etc. And this scenario seems to be a recurring problem, from time to time. <sigh> Oh well... no choice but to run another project while waiting out this disturbance in SETI's flow.

---

I DONATE money to SETI@home.... DO YOU?

I'm just slowly BOINC'ing along.

Hey... ET... you have a sister who likes earthlings?

Hmmmm.... uploading and reporting are still malfunctioning. It seems to me that someone at SETI would notice that "Results received in last hour" are consistently below normal etc, etc, etc. And this scenario seems to be a recurring problem, from time to time. <sigh> Oh well... no choice but to run another project while waiting out this disturbance in SETI's flow.

Eric already did, So Yes He's aware and hopefully something can be fixed where ever It is located at.

---

The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's

...As My habit is to detach from one project and then attach to another...

Bad habit ;-)

Why would you do that? BOINC is (very)* well capable of managing more than one project at the same time.

* removed the "very" for the newer BOINC clients ;-)

Gruß,
Gundolf

...As My habit is to detach from one project and then attach to another...

Bad habit ;-)

Why would you do that? BOINC is (very)* well capable of managing more than one project at the same time.

* removed the "very" for the newer BOINC clients ;-)

Gruß,
Gundolf

I just do and I won't say anymore on this, please don't ask.

---

The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's

I can get almost anywhere in the world on the net except Seti.

WHERE ARE THE FISH?

Apparently this may not be a Seti problem, for a change.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

I can get almost anywhere in the world on the net except Seti.

WHERE ARE THE FISH?

Apparently this may not be a Seti problem, for a change.

Periscope depth please Mr Sattler and Load all the tubes.

---

The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's

I will boot WHINE.EXE if I actually run out of work.

.... and here I thought you ran that as a service.

---

I can get almost anywhere in the world on the net except Seti.

WHERE ARE THE FISH?

Apparently this may not be a Seti problem, for a change.

Periscope depth please Mr Sattler and Load all the tubes.

Tubes are all plugged.

Got a few more day's worth of Seti to crunch. Then all bets are off.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

I can get almost anywhere in the world on the net except Seti.

WHERE ARE THE FISH?

Apparently this may not be a Seti problem, for a change.

Periscope depth please Mr Sattler and Load all the tubes.

Tubes are all plugged.

Got a few more day's worth of Seti to crunch. Then all bets are off.

Then close the doors and prepare the Nautilus for Ramming Mr Sattler. ;)

We'll sink that scalawag yet. :D

---

The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's

I can get almost anywhere in the world on the net except Seti.

WHERE ARE THE FISH?

Apparently this may not be a Seti problem, for a change.

Periscope depth please Mr Sattler and Load all the tubes.

Tubes are all plugged.

Got a few more day's worth of Seti to crunch. Then all bets are off.

Then close the doors and prepare the Nautilus for Ramming Mr Sattler. ;)

We'll sink that scalawag yet. :D

The tubes are loaded bud.

Bismark

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

The tubes are loaded bud.
Bismark

All he can think off is FISH:
pickled herring <=> in german: der Bismarckhering ;-) (not only the battleship)

__W__

---

_______________________________________________________________________________

Oh, a new error message from Seti-Server since 18:28 UTC:

"Scheduler request failed: Server returned nothing (no Headers, no data)."

Seems, someone is working around..., and that on weekend oO.

---

- Performance is not a simple linear function of the number of CPUs you throw at the problem. -

Did you really not fancy this Mark.

http://www.huffingtonpost.com/2010/02/16/worlds-strongest-beer-sco_n_463975.html

Dave

---

Well I should have done this days ago but I finally pushed the no new work on all 3 computers. I cant see getting the stray work unit when I cant upload or send any ready to report back. So when they go I will power off the old P4 till the outage is over. That old thing needs a break any way, been ruunning 24/7 since 2004 . The mac is chokeing on the few seti it has left but Milkyway is chugging along fine. As far as the i7 goes I just attached to milkyway. So we will see what 8 cores running regular opps can do for my rac there.

Boy I wish Rac would rise as fast as it falls:) So thats the view from here.

---

[/quote]

Old James

Well I should have done this days ago but I finally pushed the no new work on all 3 computers. I cant see getting the stray work unit when I cant upload or send any ready to report back. So when they go I will power off the old P4 till the outage is over. That old thing needs a break any way, been ruunning 24/7 since 2004 . The mac is chokeing on the few seti it has left but Milkyway is chugging along fine. As far as the i7 goes I just attached to milkyway. So we will see what 8 cores running regular opps can do for my rac there.

Boy I wish Rac would rise as fast as it falls:) So thats the view from here.

Use the opt apps and you will catch up to your Seti credit total in no time. Their opt app is extremely efficient and Credit/hr is very high especially on the GPU.

Well I should have done this days ago but I finally pushed the no new work on all 3 computers. I cant see getting the stray work unit when I cant upload or send any ready to report back. So when they go I will power off the old P4 till the outage is over. That old thing needs a break any way, been ruunning 24/7 since 2004 . The mac is chokeing on the few seti it has left but Milkyway is chugging along fine. As far as the i7 goes I just attached to milkyway. So we will see what 8 cores running regular opps can do for my rac there.

Boy I wish Rac would rise as fast as it falls:) So thats the view from here.

Use the opt apps and you will catch up to your Seti credit total in no time. Their opt app is extremely efficient and Credit/hr is very high especially on the GPU.

My Nvidia geforce 250 isnt good enough for milkyway. And ill run stock just to see what my 8 cores can do, When seti gets back up milkyway will get the no new tasks for this i7.

---

[/quote]

Old James

This definitely seems to be a problem that is effecting those machines with a lot of units to upload a LOT more than machines with only a few.

Take a look through the top 100. Those that are running a big number of low power machines such as This One and a few others do not seem to be worried at all. Their throughput and RAC's seem back to normal and they don't seem to have been effected by the upload problems that began last weekend, only by the server problems in the middle of last week

What does the "Database Queries/second" on the server status page represent ? I notice this is down to about half "normal"

Brodo

This definitely seems to be a problem that is effecting those machines with a lot of units to upload a LOT more than machines with only a few.

Take a look through the top 100. Those that are running a big number of low power machines such as This One and a few others do not seem to be worried at all. Their throughput and RAC's seem back to normal and they don't seem to have been effected by the upload problems that began last weekend, only by the server problems in the middle of last week

What does the "Database Queries/second" on the server status page represent ? I notice this is down to about half "normal"

Brodo

A significant answer to that comes from my successful download (at full transfer speed - 100KBytes/sec, sustained) earlier this afternoon.

One scheduler request (== one database query, to a first appoximation) got 69 tasks. I think few requests are successfully reaching Anakin - but once the full request content is accepted, there is no difficulty in processing it.

Under more normal conditions, there are a higher number of successful secheduler connections (== database queries), but each one gets fewer tasks, so download bandwidth is much the same.

I haven't looked at the scheduler (Anakin), but in Bruno's case (upload server), packets seem to be reaching it with no problem: it's Bruno itself which is issuing 'RST' packets to kill the connection. Server, not comms.