Saw this in my message log and laughed.

2012-04-11 18:57:16|SETI@home|Sending scheduler request: To fetch work. Requesting 1 seconds of work, reporting 1 completed tasks

It is inevitable that there would be a 1-second work request, but I never thought I would see it. Lowest I've seen before that was 12 seconds.

If I didn't know better, I'd swear that was an April Fools joke...

---

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

Saw this in my message log and laughed.

2012-04-11 18:57:16|SETI@home|Sending scheduler request: To fetch work. Requesting 1 seconds of work, reporting 1 completed tasks

It is inevitable that there would be a 1-second work request, but I never thought I would see it. Lowest I've seen before that was 12 seconds.

If I didn't know better, I'd swear that was an April Fools joke...

I routinely see 1 second requests on my C2D. I also keep a 2 hour cache total on my machine so I never have a large request going in. I also crunch 11 total projects 1 of which is GPU only. I never try for a large cache because I like to turn my work around very fast. If a project is down so what it will get extra work later it is all balanced out later.

---

backup projects also do 1 sec requests

---

I'm not the Pope. I don't speak Ex Cathedra!

backup projects also do 1 sec requests

So does the project initialisation request when you join a new project.

backup projects also do 1 sec requests

So does the project initialisation request when you join a new project.

it felt a bit silly to mention that ;)

---

I'm not the Pope. I don't speak Ex Cathedra!

backup projects also do 1 sec requests

So does the project initialisation request when you join a new project.

it felt a bit silly to mention that ;)

Good to know. But this is my one and only project. Was just topping-off the full 10-day cache. :D

---

Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)

Scheduler may be having issues again.
Getting "Project has no tasks available" messages, but more than usual. Add to that i've also been getting "No tasks sent" messages as well. It doesn't say why (like it does if you've reached the server side limits), just "No tasks sent".

---

Grant
Darwin NT

Getting more & more "No tasks sent" responses from the Scheduler when requesting work.

---

Grant
Darwin NT

57 SETI@home 13.04.2012 11:55:31 Requesting new tasks for GPU
62 SETI@home 13.04.2012 11:57:49 Scheduler request completed: got 0 new tasks
63 SETI@home 13.04.2012 11:57:49 Message from server: No tasks sent
64 SETI@home 13.04.2012 11:57:49 Message from server: No tasks are available for Astropulse v505
65 SETI@home 13.04.2012 11:57:49 Message from server: No tasks are available for SETI@home v7
66 SETI@home 13.04.2012 11:57:49 Message from server: No tasks are available for AstroPulse v6
73 13.04.2012 11:58:38 Internet access OK - project servers may be temporarily down.

Panic Mode On, Server Problems!

---

Getting more & more "No tasks sent" responses from the Scheduler when requesting work.

IIRC, you get the "No tasks sent" message if there are tasks available but not for the type of your request (only VLARs if you ask for GPU tasks).

Gruß,
Gundolf

57 SETI@home 13.04.2012 11:55:31 Requesting new tasks for GPU
62 SETI@home 13.04.2012 11:57:49 Scheduler request completed: got 0 new tasks
63 SETI@home 13.04.2012 11:57:49 Message from server: No tasks sent
64 SETI@home 13.04.2012 11:57:49 Message from server: No tasks are available for Astropulse v505
65 SETI@home 13.04.2012 11:57:49 Message from server: No tasks are available for SETI@home v7
66 SETI@home 13.04.2012 11:57:49 Message from server: No tasks are available for AstroPulse v6
73 13.04.2012 11:58:38 Internet access OK - project servers may be temporarily down.

Panic Mode On, Server Problems!

[NVidia] GPU request, all [v6] MB tasks left in the feeder VLAR, so can't be sent. 'sent from other apps' enabled, so it goes through the list of other apps, but no such tasks in the feeder either.

---

I'm not the Pope. I don't speak Ex Cathedra!

Yup, something is faltering, caches dropping, Crickets are getting quieter.

---

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

57 SETI@home 13.04.2012 11:55:31 Requesting new tasks for GPU
62 SETI@home 13.04.2012 11:57:49 Scheduler request completed: got 0 new tasks
63 SETI@home 13.04.2012 11:57:49 Message from server: No tasks sent
64 SETI@home 13.04.2012 11:57:49 Message from server: No tasks are available for Astropulse v505
65 SETI@home 13.04.2012 11:57:49 Message from server: No tasks are available for SETI@home v7
66 SETI@home 13.04.2012 11:57:49 Message from server: No tasks are available for AstroPulse v6
73 13.04.2012 11:58:38 Internet access OK - project servers may be temporarily down.

Panic Mode On, Server Problems!

[NVidia] GPU request, all [v6] MB tasks left in the feeder VLAR, so can't be sent. 'sent from other apps' enabled, so it goes through the list of other apps, but no such tasks in the feeder either.

Could the feeder's lots be increased from 100 to 1,000? Or more? That way, there would be a greater chance of what an individual host requests being there.

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

Could the feeder's lots be increased from 100 to 1,000? Or more? That way, there would be a greater chance of what an individual host requests being there.

I think it's been mentioned that when everything is working up to par, the current feeder setup is quite capable of saturating the bandwidth currently available, so increasing feeder activity by whatever means would not really increase the availability of work.
Even if you got tasks assigned, you wouldn't be able to effectively download it, and that would stop your Boinc from even asking for more due to backoffs. So the end result would be the same as getting nothing on a work request.

If more bandwidth is made available in the future, then I would agree that some enhancement of the feeder/scheduler process might be in the cards as well.

---

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

Could the feeder's lots be increased from 100 to 1,000? Or more? That way, there would be a greater chance of what an individual host requests being there.

I think it's been mentioned that when everything is working up to par, the current feeder setup is quite capable of saturating the bandwidth currently available, so increasing feeder activity by whatever means would not really increase the availability of work.
Even if you got tasks assigned, you wouldn't be able to effectively download it, and that would stop your Boinc from even asking for more due to backoffs. So the end result would be the same as getting nothing on a work request.

If more bandwidth is made available in the future, then I would agree that some enhancement of the feeder/scheduler process might be in the cards as well.

IIRC recently they did a test and determined the current hardware could handle 0.5Gb connection. That is 5X our current pipe.

---

Could the feeder's lots be increased from 100 to 1,000? Or more? That way, there would be a greater chance of what an individual host requests being there.

I think it's been mentioned that when everything is working up to par, the current feeder setup is quite capable of saturating the bandwidth currently available, so increasing feeder activity by whatever means would not really increase the availability of work.
Even if you got tasks assigned, you wouldn't be able to effectively download it, and that would stop your Boinc from even asking for more due to backoffs. So the end result would be the same as getting nothing on a work request.

If more bandwidth is made available in the future, then I would agree that some enhancement of the feeder/scheduler process might be in the cards as well.

Indeed the feeder can do so. Like a few weeks ago when there were do many active downloads during maintenance the bandwidth stayed pegged during the outage.

With just normal AR MB tasks going out the graph does normally tend to be wavy instead of pegged. However with task limits, transfer issues, shorties, and/or AP tasks going out the bandwidth does normally stay pegged. The current graph does look like there would be something wrong under those circumstances, but it is probably just normal AR work going out without any network issues. So in that case the feeder does lag behind.

The feeder queue could probably be made more dynamic to adjust to the bandwidth in use. Something tells me that would be needlessly complicated & probably not work as expected.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

Could the feeder's lots be increased from 100 to 1,000? Or more? That way, there would be a greater chance of what an individual host requests being there.

I think it's been mentioned that when everything is working up to par, the current feeder setup is quite capable of saturating the bandwidth currently available, so increasing feeder activity by whatever means would not really increase the availability of work.
Even if you got tasks assigned, you wouldn't be able to effectively download it, and that would stop your Boinc from even asking for more due to backoffs. So the end result would be the same as getting nothing on a work request.

There would at least be fewer scheduler requests mixed into the total traffic in the pipe, but I suppose that's a pretty insignificant part of the total.

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

I know, some people will not like me anymore after they have read following remarks, but I have to write them down:
- Download works at a quite unusual high speed of appr. 150 kB / sec. This value is one of the highest I've ever seen!
- Bandwith is not maxed out as I can see.
- And all this is only possible when all the AP splitters are not running! Happened several times in the last few months.
- Could there be a relationship between AP splitters not running and speed of download?
- If yes, why not try to separate AP totally from MB? I know, this would take some time and (even more) money. But why not at least discuss it.

Glad to hear multiple opinions of fellow crunchers!

Rolf

Over the last week when boinc has asked for a task it's got one, and the number of re-tries has plummeted (down to one or two!). So I'm complaining about not having much to complain about :)

---

Member of the People Encouraging Niceness In Society club.

Over the last week when boinc has asked for a task it's got one, and the number of re-tries has plummeted (down to one or two!). So I'm complaining about not having much to complain about :)

Isn't a real bugger when there's nothing to complain and/or panic about (except for those limits)?

Cheers.

---