BOINC 6.6.11 is out. As stated in other threads this is a Development version. The change log is below.

6.6.11

- client: if a project is at max backoff for a resource, stop accumulating debt if it's at or around zero.

This prevents other projects from being driven unboundedly negative.

- client: if the number of overworked projects exceeds the number of device instances, clear debts; this indicates that an earlier client was buggy and produced bad debt values.

- client: new work-fetch policy:

1) if an instance is idle, get work from highest-debt project, even if it's overworked.

2) if resource has a shortfall, get work from highest-debt non-overworked project

3) if there's a fetchable non-overworked project with no runnable jobs, get from the highest-debt one
(each step is done first for GPU, then CPU)

Clause 3) is new.

It will cause the client to get jobs for as many projects as possible, even if there is no shortfall. This is necessary to make the notion of "overworked" meaningful (otherwise, any project with long jobs can become overworked). It also maintains as much variety as possible (like pre-6.6 clients).

Also (small bug fix) if a project is overworked for resource R, request work for R only in case 1).

- client: remove the "debt repair" mechanism added earlier today.

There are situations where multiple projects can legitimately have large negative LTD on a uniprocessor. Instead...

- client: add <zero_debts> option to cc_config.xml

- client: fix bug that can cause no GPU jobs to be scheduled even when a GPU is free

- client: tweak the LTD calculation so that the largest debt always tends towards zero

- client: print message if downloaded file has wrong size

- client: associate file xfer messages with a project; fixes #848

- client: more instances of showing project with message. Fixes #848

- client: fix work-fetch bug that caused infinite fetch; cleanup/reorganization of work fetch logic

- client: work-fetch bug fix: if we're fetching work for a starved project, it most have no runnable jobs for ANY resource.

- client: work-fetch bug fix: when setting requests in the shortfall case, don't request anything if project is backed off or overworked for the resource.

- scheduler: include driver version in the CUDA description string storing in the database;

- WINSCR: Fix the BOINC text displayed is the screensaver control panel applet

- WINSCR: Get rid of the 'Up to 5 second wait' for the data thread to terminate. Force terminate the thread and clean up the graphics app in a clean-up routine. Forcing the user to wait when they return to the system isn't a good idea.

- WINSCR: Fix screensaver issue where the screensaver locks up when the graphics application deadlocks for one reason or another.

- WINSCR: Clean-up some of the thread cleanup logic.

---

BOINC blog

Seems that 6.6.11 doesn't care whether I hca 'No new tasks' flag on the project..
It just downloaded GPUgrid WU, but I definitely have it as 'no new tasks'.

---

--
Get up, stand up! Don\'t give up the fight!
Credits will make everybody feel high! ;-)

That is a scheduler problem, not a client problem. Easily tested by attaching to any other project and putting it on NNT. You'll notice it'll work.

Normally it goes like this:
Your BOINC says "hi, I am here but not asking for work" to scheduler at project. Scheduler says "hi there, nice to have you around."
Since your BOINC isn't asking for work, scheduler does nothing.

In the case of GPUGrid and for a while Seti Beta as well, it went:
BOINC says "hi, I am here but not asking for work" to scheduler at project. Scheduler says "hi there, nice to have you around and here, have some work anyway!"

---

I have it for Einstein@home as well.
Furthermore, I have suspended the GPUgrid, and still getting new task everytime I wake up my PC (Vista 32bit).

Didn't have that behaviour with 6.6.9

---

--
Get up, stand up! Don\'t give up the fight!
Credits will make everybody feel high! ;-)

Some random behaviour:
--
26/02/2009 20:24:08 GPUGRID Sending scheduler request: Requested by user.
26/02/2009 20:24:08 GPUGRID Reporting 4 completed tasks, not requesting new tasks
26/02/2009 20:24:14 GPUGRID Scheduler request completed: got 1 new tasks
--

---

--
Get up, stand up! Don\'t give up the fight!
Credits will make everybody feel high! ;-)

If you want to show what it's doing, enable the <work_fetch_debug>, <sched_op_debug> and <dcf_debug> flags in cc_config.xml and most critically, sent that output to the BOINC Alpha email list. This list will need registration.

---

I'm still getting Seti Beta tasks with NNT selected, this time on Boinc 6.6.11,
and couldn't get any tasks for Seti Main even with all other projects suspended,
so enabled those flags and will have a look at it tomorrow, and send files.
Had to downgrade to 6.4.6 to get some work, before upgrading back to 6.6.11.

Claggy

@ Claggy,

Please see my note at Beta about additional data you could try and preserve.

I'm still getting Seti Beta tasks with NNT selected, this time on Boinc 6.6.11,
and couldn't get any tasks for Seti Main even with all other projects suspended,
so enabled those flags and will have a look at it tomorrow, and send files.
Had to downgrade to 6.4.6 to get some work, before upgrading back to 6.6.11.

Claggy

I recall Dr A saying something on the boinc_alpha mailing list about fixing this in 6.6.12.

---

BOINC blog

I'm still getting Seti Beta tasks with NNT selected, this time on Boinc 6.6.11,
and couldn't get any tasks for Seti Main even with all other projects suspended,
so enabled those flags and will have a look at it tomorrow, and send files.
Had to downgrade to 6.4.6 to get some work, before upgrading back to 6.6.11.

Claggy

I recall Dr A saying something on the boinc_alpha mailing list about fixing this in 6.6.12.

He wrote on the BOINC_dev list that it would be fixed in v6.6.9 (18 February 2009 23:17 UTC).

But David is fixated on this being a client bug. There were client bugs, but I think they've been ironed out by now. The possibility of a server bug remains, but is not being considered by the 'powers that be'.

That's why I suggested Claggy captures the sched_request and sched_reply files too.

At least I'm not the only one.

2/28/2009 5:01:07 AM SETI@home Sending scheduler request: Requested by user.
2/28/2009 5:01:07 AM SETI@home Reporting 4 completed tasks, not requesting new tasks
2/28/2009 5:01:18 AM SETI@home Scheduler request completed: got 16 new tasks

He wrote on the BOINC_dev list that it would be fixed in v6.6.9 (18 February 2009 23:17 UTC).

But David is fixated on this being a client bug. There were client bugs, but I think they've been ironed out by now. The possibility of a server bug remains, but is not being considered by the 'powers that be'.

That's why I suggested Claggy captures the sched_request and sched_reply files too.

Well, he did find another bug still present in v6.6.11, there client claims it's not requesting work, but in reality does request for work, and this should hopefully be fixed in v6.6.12.

It's of course possible there also is a server-side bug, but, if there's no bug-reports from users of v6.2.xx and earlier, it's a fairly good chance it is client-side.

---

"I make so many mistakes. But then just think of all the mistakes I don't make, although I might."

He wrote on the BOINC_dev list that it would be fixed in v6.6.9 (18 February 2009 23:17 UTC).

But David is fixated on this being a client bug. There were client bugs, but I think they've been ironed out by now. The possibility of a server bug remains, but is not being considered by the 'powers that be'.

That's why I suggested Claggy captures the sched_request and sched_reply files too.

Well, he did find another bug still present in v6.6.11, there client claims it's not requesting work, but in reality does request for work, and this should hopefully be fixed in v6.6.12.

It's of course possible there also is a server-side bug, but, if there's no bug-reports from users of v6.2.xx and earlier, it's a fairly good chance it is client-side.

Ingleside,

Do you remember my Work fetch anomaly thread - you contributed to it.

That was a v5.10.13 client: requested 69 seconds, got 12 days. I doubt there was a 'hidden' CUDA work fetch in that version.

Ingleside,

Do you remember my Work fetch anomaly thread - you contributed to it.

That was a v5.10.13 client: requested 69 seconds, got 12 days. I doubt there was a 'hidden' CUDA work fetch in that version.

There was some buggy scheduling-server-code also, but AFAIK these bugs was fixed back in January.

But, one or more bugs fixed back in January doesn't mean there aren't any more bugs hidden-away server-side.

---

"I make so many mistakes. But then just think of all the mistakes I don't make, although I might."

Switched over my Cuda hosts this morning. Everything working fine and dandy :)

---

"The most exciting phrase to hear in science, the one that heralds new discoveries, is not Eureka! (I found it!) but rather, 'hmm... that's funny...'" -- Isaac Asimov

I'm still getting Seti Beta tasks with NNT selected, this time on Boinc 6.6.11,
and couldn't get any tasks for Seti Main even with all other projects suspended,
so enabled those flags and will have a look at it tomorrow, and send files.
Had to downgrade to 6.4.6 to get some work, before upgrading back to 6.6.11.

Claggy

I recall Dr A saying something on the boinc_alpha mailing list about fixing this in 6.6.12.

He wrote on the BOINC_dev list that it would be fixed in v6.6.9 (18 February 2009 23:17 UTC).

But David is fixated on this being a client bug. There were client bugs, but I think they've been ironed out by now. The possibility of a server bug remains, but is not being considered by the 'powers that be'.

That's why I suggested Claggy captures the sched_request and sched_reply files too.

Here is the exerpt from boinc_alpha mailing list. I will qualify it though that it relates to GPU project (which claggy may not be running). As you say there may be further bugs on the server side.

Email addresses blanked out for spam protection.

Date: Fri, 27 Feb 2009 10:59:02 -0800
From: David Anderson <davea@
Subject: Re: [boinc_alpha] 6.6.11: NNW does not work for GPU project
To: BOINC Alpha Mailing List <boinc_alpha@
Message-ID: <49A837F6.3020405@
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

I fixed this problem (I hope).
It will appear in 6.6.12
- David

---

BOINC blog

Yes, I'm running Cuda on Seti Main with Raistmer's V7 app at the moment,
I've been up to V9 and V10, and down to V7 a couple of times,
I just can't seem to get any Cuda work from Seti Main with 6.6.11, it asks for it, but doesn't get any,
while i get Cuda work on Beta with nnt selected.
I even wondered if the nnt buttons were cross connected, but no.

Claggy

There are reports at Lunatics that scheduler requests for CUDA work (using BOINC clients that request CPU and CUDA work separately - I'm not sure how far you have to go up the v6.6.x ladder for that) are being met with 86400 second backoffs - not entirely clear whether this is a server demand, or a client reaction, but it seems to be correlated with receiving no work. Older clients, which just request work without splitting it down into different types, are getting work OK. You probably need to activate "Work Fetch Debug" to see all the relevant messages.

If you want to show what it's doing, enable the <work_fetch_debug>, <sched_op_debug> and <dcf_debug> flags in cc_config.xml and most critically, sent that output to the BOINC Alpha email list. This list will need registration.

You mean the messages in 'stdoutdae(.old)'?

Or create the '_debug>' entries in cc_config an new file in the BOINC folder?

---

---

"The most exciting phrase to hear in science, the one that heralds new discoveries, is not Eureka! (I found it!) but rather, 'hmm... that's funny...'" -- Isaac Asimov