scarecrow's script says we need more complaining:D

please lift the limits, thank you.

---

looks like the limits have been adjusted, graphs are pegged again.

thank you

---

looks like the limits have been adjusted, graphs are pegged again.

thank you

That spike started off as an Astropulse spike, i got 12 of them, and the Astropulse ready to send dropped by 900 Wu's,

Claggy

Uh-Oh. Cricket graphs not available. Are we still UL/DL-ing?
(Edit) I just tried UL and DL. Still working fine, but no Cricket graph to see it.

Uh-Oh. Cricket graphs not available. Are we still UL/DL-ing?

7/18/2010 12:59:17 PM SETI@home Sending scheduler request: To fetch work.
7/18/2010 12:59:17 PM SETI@home Reporting 8 completed tasks, requesting new tasks
7/18/2010 12:59:27 PM SETI@home Scheduler request completed: got 22 new tasks


Yes we are. :-) Although downloads are VERY slow.

---

please lift the limits, thank you.

---

please lift the limits, thank you.

If you have not enough SETI@home WUs, what about SETI@home Beta Test WUs? ;-)

'Test with us - VLAR WUs not to GPUs..'

It's little bit sad, that only a few members posted their experiences at the SETI@home Beta Test forum..
No one else have interest to have this function at SETI@home?
It needed to be tested at BETA first..

---

BTW.

AFAIK, SETI@home Beta Test WUs are the same like SETI@home WUs. They are also for the science.
SETI@home Beta Test WUs are not 'test-WUs'.

---

I have removed the limits for this run on the assumption that, after a few hours of being pegged, bandwidth usage will fall well below max. let's see what happens.

---

That's great Jeff....Now, if only we could get some work units! :-)

7/18/2010 6:03:20 PM SETI@home Scheduler request completed: got 0 new tasks
7/18/2010 6:03:20 PM SETI@home Message from server: Project has no tasks available

Never mind me, there's no problem here...

7/18/2010 6:07:55 PM SETI@home Scheduler request completed: got 4 new tasks

---

PROUD MEMBER OF Team Starfire World BOINC

19/07/2010 00:09:44 SETI@home Requesting new tasks for CPU and GPU
19/07/2010 00:10:01 SETI@home Scheduler request completed: got 0 new tasks
19/07/2010 00:10:01 SETI@home Message from server: Project has no tasks available

Posting this since it seems I (we) often get new WUs after posting... :D

Thanks, Jeff. Your timing this weekend on allowing work through the pipeline has been great! We are cooking with gas!!!

Jeff Bakle

---

19/07/2010 00:09:44 SETI@home Requesting new tasks for CPU and GPU
19/07/2010 00:10:01 SETI@home Scheduler request completed: got 0 new tasks
19/07/2010 00:10:01 SETI@home Message from server: Project has no tasks available

Posting this since it seems I (we) often get new WUs after posting... :D

Well one at least, and it was an AP.

19/07/2010 00:15:23 SETI@home Sending scheduler request: To fetch work.
19/07/2010 00:15:23 SETI@home Reporting 1 completed tasks, requesting new tasks for CPU
19/07/2010 00:15:27 SETI@home Scheduler request completed: got 1 new tasks

Edit:

19/07/2010 00:21:48 SETI@home Sending scheduler request: To fetch work.
19/07/2010 00:21:48 SETI@home Requesting new tasks for CPU and GPU
19/07/2010 00:21:57 SETI@home Scheduler request completed: got 16 new tasks

I have removed the limits for this run on the assumption that, after a few hours of being pegged, bandwidth usage will fall well below max. let's see what happens.

Jeff

While where you post really does not matter... I would suggest that after the outage you post a new thread in the NC and make it Sticky and then lock it. Then You announcments are visible for everyone and easy to find. By locking it there will be no clutter... Then Users can reference the correct post and go about other conversation.

At the end you can just make the thread Non Sticky and it drifts into the archives...

If there is something happening that you do want interaction with the users, make it sticky at the top and the people in NC that do have comments (Good or Bad) will eat it up. YOu could do a short note on expect results due to the changes.

You could even "CutnPaste" in the Tech New or News and then lock it. You only have one place to look.

Yes most will complain elsewhere, and then get into the serious discussion here.

Regards

Pappa

---

Please consider a Donation to the Seti Project.

7/18/2010 7:23:46 PM	SETI@home	Reporting 1 completed tasks, requesting new tasks for CPU
7/18/2010 7:23:51 PM	SETI@home	Scheduler request completed: got 0 new tasks
7/18/2010 7:23:51 PM	SETI@home	Message from server: Project has no tasks available

why do i get no task available when there are "Results ready to send 298,288"?

---

why do i get no task available when there are "Results ready to send 298,288"?

There are three possible answers:

1. The info on the server status page may be several minutes old, and a lot of results could have been sent in that time

2. I think "no tasks available" is a standard answer when you can't complete a connection to the servers - you will get this when all the server connections are already in use for example, or if there are other connection problems

3. It is an evil plot by the Masters of S@H

---

why do i get no task available when there are "Results ready to send 298,288"?

There are three possible answers:

1. The info on the server status page may be several minutes old, and a lot of results could have been sent in that time

2. I think "no tasks available" is a standard answer when you can't complete a connection to the servers - you will get this when all the server connections are already in use for example, or if there are other connection problems

3. It is an evil plot by the Masters of S@H

The "No Tasks Available" message comes from the Scheduler - your request has to reach the Scheduler server and be processed to receive that answer. If all the Server connections are full, or the Servers are down, you get the "Cannot Connect to Servers" message.

What most likely happened is your request hit the Scheduler when the Download Feeder was empty.

Downloads are NOT a continuous feed. IIRC, the download feeder process has 100 WU slots, filled with a mix of S@H Enhanced, Astropulse, and CUDA tasks. The feeder refills every 5-6 seconds. If your request hits the Scheduler during the portion of that 6-second cycle when the Feeder has no WUs of the type your client is requesting, you get the "No Tasks Available" message, no matter how many "WUs Ready to Send" the Server Status page shows.

When this happens, you have two choices:

1) (Preferred method) Relax, and allow the automatic back-off feature in your BOINC Manager to try again after a short wait. It will keep trying until it gets through and receives new WUs (or you tell it to stop).

2) Repeatedly click the "Update" button on your BOINC Manager, hoping you will get lucky and catch the next cycle when it has WUs you can use. While this may hasten the time when you get WUs, it also increases Server and Database loading. During peak traffic times, if a LOT of folks do this, it can cause slowdowns at best, and crashes at worst.

Hope that helps.

Donald
Infernal Optimist / Submariner, retired

and once again, I have two tasks I'm working on, and none whatsoever buffered...

this is getting ridiculous

19/07/2010 2:24:42 PM SETI@home Sending scheduler request: To fetch work.
19/07/2010 2:24:42 PM SETI@home Requesting new tasks
19/07/2010 2:24:48 PM SETI@home Scheduler request completed: got 0 new tasks
19/07/2010 2:24:48 PM SETI@home Message from server: Project has no tasks available


PS: I have an ATI vid card, so I am running with GPU disabled... since there's no point requesting GPU tasks that I'll never get.

will have to see in around 13 mins if I actually get another WU, or if I'll just completely run out.


Edit:

OMG .. I finally got a chunk of tasks... 12 in fact

Just had a look & it appears my cache is full. On the Cricket graphs downloads are still going flat out, but inbound is slowly tapering off so it looks like most of the other caches have filled up or are close to it.

---

Grant
Darwin NT

While where you post really does not matter... I would suggest that after the outage you post a new thread in the NC and make it Sticky and then lock it. Then You announcments are visible for everyone and easy to find. By locking it there will be no clutter... Then Users can reference the correct post and go about other conversation.

At the end you can just make the thread Non Sticky and it drifts into the archives...

If there is something happening that you do want interaction with the users, make it sticky at the top and the people in NC that do have comments (Good or Bad) will eat it up. YOu could do a short note on expect results due to the changes.

You could even "CutnPaste" in the Tech New or News and then lock it. You only have one place to look.

Yes most will complain elsewhere, and then get into the serious discussion here.

Regards

Pappa

Great idea. Thanks!

Jeff

---