The return message to client is "NO WORK FROM PROJECT"...... Boinc 4.19, seti 4.18, win ME

---

That is because the page is made and displayed for "x" minutes... so, by the time you see it the 15 are gone already ...

---

That is because the page is made and displayed for "x" minutes... so, by the time you see it the 15 are gone already ...

Not quite, it still reads "27" and I read it 40 mins ago est. No matter I have a file retrived in last 10 mins and page still reads 27......

---

As Paul said, the status page is not a source for up-to-the-second information. At the time the page was generated there were 27 ready to send. All of those probably went out within seconds but the splitter was busy inserting more at the same time. Wether or not there is a work unit to send at the very instant YOU connect is just luck of the draw when the number is so low. Join a 2nd project to keep busy. Oh. I digress as well. That must be contagious, Paul!

---

A member of The Knights Who Say NI!
For rankings, history graphs and more, check out:
My BOINC stats site

Lets look at the download problem this way. Let's say you are working on a production line. It's your job to keep up your parts so the output production doesn't suffer. Normally you don't break much of a sweat keeping up. Then the management orders you to take a two day suspension. Then when you get back to work, man is the work piled up. You can only do so much, but the demand at the end isn't letting up. You have two days worth of work to catch up on. You work as hard as you can, but even at your best you can't keep up, but you come close to keeping the line fed. Sometimes you have 16 parts waiting for pick up and sometimes 27 and all the numbers in between. Then the forklift comes around take what you got and your back to nothing to feed the line with.

The splitters are splitting, but the demand after this long outtage just isn't letting up any time soon. It makes the WUs, but are snatched right up.

does this make it clearer?

tony

We may also be seeing another side-effect of the last outage and problems since. I am going to suggest that many people have upped the size of their queues. we will be able to see that when the system stabalizes ... we had a million and change in-flight before and may see that number rise. I don't normally track that number so, it is hard to say if my memory serves correctly.

To help out, I have left most of my machines with SETI@Home suspended for the time being. Heck, I still have a ton of work in pending state and with me working on the material for installing optimized clients well, I don't want there to be that much work on the machines (of course I have a couple machines with work lined up so I will see if they barf when I do the installs).

As a digression, it looks like Einstein@Home is working on their own optimized client applications. With the one for OS-X using the altevec when possible... WAY cool ...

---