I would say this should be looked into, just in case WUs are really getting lost. Can someone please pass the word up the food chain?

Look at your task list, and filter it to 'pending' only.

Problem is, my task list is huge, as my RAC is roughly 45K/day. Looking through it would be a killer. I don't have a machine that (ideally) does a handfull of WUs/day. Didn't someone at the top of the thread report this "effect" from such a machine?

Since I am one of the little guys I took a look. I have 191 tasks pending through the task page while the pending credit page only lists 84 of them. The last unit on the pending credit page was reported March 8, 2011, at 12:34:55 UTC. The next one doesn't show up even though I have plenty of new work returned since.
I'm not overly concerned with it other than not knowing how much potential credit I have waiting which really doesn't amount to a hill of beans.
Dan

---

I would say this should be looked into, just in case WUs are really getting lost. Can someone please pass the word up the food chain?

Look at your task list, and filter it to 'pending' only.

Problem is, my task list is huge, as my RAC is roughly 45K/day. Looking through it would be a killer. I don't have a machine that (ideally) does a handfull of WUs/day. Didn't someone at the top of the thread report this "effect" from such a machine?

Since I am one of the little guys I took a look. I have 191 tasks pending through the task page while the pending credit page only lists 84 of them. The last unit on the pending credit page was reported March 8, 2011, at 12:34:55 UTC. The next one doesn't show up even though I have plenty of new work returned since.
I'm not overly concerned with it other than not knowing how much potential credit I have waiting which really doesn't amount to a hill of beans.
Dan

IMO: Hill of beans value > pending tasks value.

---

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

Oh dear. I'm always surprised how few computer users understand how a database works. -BeNt- thinks that we're looking at the output of a script (msg 1088168), and HAL9000 thinks it's its stored in a table (1085736).

In fact, all tasks are held in one big (giant) table: every user, every task - of the order of ten million plus of them. Every time you click on that link, you're asking Carolyn (or possibly Jocelyn) to look through all ten million lines, and find out:
a) which are mine,
b) which I've done,
c) and which are still pending.

It's called a query, the code is in http://boinc.berkeley.edu/trac/browser/trunk/boinc/html/user/pending.php (line 67), and it's very simple:

$results = BoincResult::enum("userid=$user->id AND (validate_state=0 OR validate_state=4) AND claimed_credit > 0");

Userid is obvious
I doubt they've changed validate_state meanings
so I suspect they're not storing claimed_credit any longer (or storing 0): we had a discussion recently about the disconnection between the old "claimed" credit based on flop-counting (used by the project until June 2010), and "CreditNew" used since then. Somebody's probably just turned off the last vestiges of the old system.

Oh dear. I'm always surprised how few computer users understand how a database works. -BeNt- thinks that we're looking at the output of a script (msg 1088168), and HAL9000 thinks it's its stored in a table (1085736).

In fact, all tasks are held in one big (giant) table: every user, every task - of the order of ten million plus of them. Every time you click on that link, you're asking Carolyn (or possibly Jocelyn) to look through all ten million lines, and find out:
a) which are mine,
b) which I've done,
c) and which are still pending.

It's called a query, the code is in http://boinc.berkeley.edu/trac/browser/trunk/boinc/html/user/pending.php (line 67), and it's very simple:

$results = BoincResult::enum("userid=$user->id AND (validate_state=0 OR validate_state=4) AND claimed_credit > 0");

Userid is obvious
I doubt they've changed validate_state meanings
so I suspect they're not storing claimed_credit any longer (or storing 0): we had a discussion recently about the disconnection between the old "claimed" credit based on flop-counting (used by the project until June 2010), and "CreditNew" used since then. Somebody's probably just turned off the last vestiges of the old system.

Yes for some reason I was thinking this data was stored in a separate table for some reason. From many postings from Matt it was clear that the user pending list was something intensive on the db. So they would sometimes shut it off after Tuesday maintenance until things recovered.

If no value is being saved for 'claimed_credit' any longer that makes sense as to why this is operating this way. However, I think it would have made more sense to disable the function instead of leaving it enabled, but that is a UI issue instead of functionality. So would go to the lower end of the task list.

I never found the pending credit value useful, but being able to view them all in one list, more than 20 at a time, is nice. Everything is still better then the S@H Classic days. When the web stats were generated once per day if we were lucky.

---

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

Oh dear. I'm always surprised how few computer users understand how a database works. -BeNt- thinks that we're looking at the output of a script (msg 1088168), and HAL9000 thinks it's its stored in a table (1085736).

In fact, all tasks are held in one big (giant) table: every user, every task - of the order of ten million plus of them. Every time you click on that link, you're asking Carolyn (or possibly Jocelyn) to look through all ten million lines, and find out:
a) which are mine,
b) which I've done,
c) and which are still pending.

It's called a query, the code is in http://boinc.berkeley.edu/trac/browser/trunk/boinc/html/user/pending.php (line 67), and it's very simple:

$results = BoincResult::enum("userid=$user->id AND (validate_state=0 OR validate_state=4) AND claimed_credit > 0");

Userid is obvious
I doubt they've changed validate_state meanings
so I suspect they're not storing claimed_credit any longer (or storing 0): we had a discussion recently about the disconnection between the old "claimed" credit based on flop-counting (used by the project until June 2010), and "CreditNew" used since then. Somebody's probably just turned off the last vestiges of the old system.

And where do you put that query.....in a script. Commonly a PHP script.....asp....java....oh boy! And you just assumed that HAL9000 was wrong by saying it's stored in a table....then you turn around and say it's stored in a table. rofl.

I would suggest you not make your rebuttals person, you only make yourself look foolish. The query scans the database finding the appropriate tables to publish for each query. The script holds the query 'code' in which is dynamically generated upon page loading and requesting values in which the script then looks through the tables(and the fields there of) of a database or even multiple databases and publishes the results out into a readable, understandable, and communicable human read format....commonly called a web page. But you know everyone isn't as smart as you, some of us, read a lot of us, have worked in these fields for quite a while. IBM informix is nice, relational database classes are not missed.

---

Traveling through space at ~67,000mph!

Richard Haselgrove wrote:

...

$results = BoincResult::enum("userid=$user->id AND (validate_state=0 OR validate_state=4) AND claimed_credit > 0");

Userid is obvious
I doubt they've changed validate_state meanings
so I suspect they're not storing claimed_credit any longer (or storing 0)
...

Correct analysis, in changeset [trac]changeset:23118[/trac] the code to put a value for claimed_credit into the Result record was removed from sched_result.cpp .

                                                                   Joe

Richard Haselgrove wrote:

...

$results = BoincResult::enum("userid=$user->id AND (validate_state=0 OR validate_state=4) AND claimed_credit > 0");

Userid is obvious
I doubt they've changed validate_state meanings
so I suspect they're not storing claimed_credit any longer (or storing 0)
...

Correct analysis, in changeset [trac]changeset:23118[/trac] the code to put a value for claimed_credit into the Result record was removed from sched_result.cpp .

                                                                   Joe

Would you happen to know if the plan is to completely depreciate the current list that shows pending tasks, or if there is some kind of change planned to give it a reason to be?

---

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

Would you happen to know if the plan is to completely depreciate the current list that shows pending tasks, or if there is some kind of change planned to give it a reason to be?

No, Dr. Anderson hasn't made any comment about the next step AFAIK.

                                                                   Joe

Would you happen to know if the plan is to completely depreciate the current list that shows pending tasks, or if there is some kind of change planned to give it a reason to be?

No, Dr. Anderson hasn't made any comment about the next step AFAIK.

                                                                   Joe

Soooo...
We don't know if this is a plan by DA to keep us in the dark about our pending credits, as he hid the claimed credits....
Or if this an artifact of some transitional work by the Seti staff?

---

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

Would you happen to know if the plan is to completely depreciate the current list that shows pending tasks, or if there is some kind of change planned to give it a reason to be?

No, Dr. Anderson hasn't made any comment about the next step AFAIK.

                                                                   Joe

Soooo...
We don't know if this is a plan by DA to keep us in the dark about our pending credits, as he hid the claimed credits....
Or if this an artifact of some transitional work by the Seti staff?

With the new credit system I don't think claimed credit really exists any longer. I am sure they would leave the code in to calculate it the old way, but that leads to problems of "What is going on? I had 150 claimed credit, but I was only granted 10!!!!"

---

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

Would you happen to know if the plan is to completely depreciate the current list that shows pending tasks, or if there is some kind of change planned to give it a reason to be?

No, Dr. Anderson hasn't made any comment about the next step AFAIK.

                                                                   Joe

Soooo...
We don't know if this is a plan by DA to keep us in the dark about our pending credits, as he hid the claimed credits....
Or if this an artifact of some transitional work by the Seti staff?

With the new credit system I don't think claimed credit really exists any longer. I am sure they would leave the code in to calculate it the old way, but that leads to problems of "What is going on? I had 150 claimed credit, but I was only granted 10!!!!"

Well, I believe that claimed credits were taken away from our view before the 'new credit' system went into play.

But, that is not really the issue at hand.

I'd would just like to understand what is going on with the pending credits right now.

Meow.

---

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

Very strange that Dr. Anderson doesn't give an explanation.

---

Goodbye Seti Classic

Very strange that Dr. Anderson doesn't give an explanation.

Dr. Anderson normally doesn't spend time reading these forums, nor even the BOINC forums.

My guess is he did do a statistical analysis of the CreditNew system, but with insufficiently varied input data. The overall credit granted for all users combined seems to not have been badly affected, it's just these large variations for individual hosts which are bothersome.

                                                                 Joe

Anyway: they are back.
Thanks

---

Goodbye Seti Classic

Same problem here: In the "pending" list there are only 8 work units, however in the tasks lists -> pending there are more than 330(!) pending!

---

Same problem here: In the "pending" list there are only 8 work units, however in the tasks lists -> pending there are more than 330(!) pending!

Same answer. It applies to everybody, they're not picking on you specially.

Same problem here: In the "pending" list there are only 8 work units, however in the tasks lists -> pending there are more than 330(!) pending!

Same answer. It applies to everybody, they're not picking on you specially.

Richard,

I pointed that out because the last message is 3 weeks old and another user posted "they are back.". Well, mine are'nt back.

---

Same problem here: In the "pending" list there are only 8 work units, however in the tasks lists -> pending there are more than 330(!) pending!

Same answer. It applies to everybody, they're not picking on you specially.

Richard,

I pointed that out because the last message is 3 weeks old and another user posted "they are back.". Well, mine are'nt back.

Nor are mine, or anyone else's, I expect. I've no idea what XP_Freak was talking about.

I think Joe's answer is definitive, and will remain so until a code-watcher spots something new.

Same problem here: In the "pending" list there are only 8 work units, however in the tasks lists -> pending there are more than 330(!) pending!

Same answer. It applies to everybody, they're not picking on you specially.

Richard,

I pointed that out because the last message is 3 weeks old and another user posted "they are back.". Well, mine are'nt back.

Nor are mine, or anyone else's, I expect. I've no idea what XP_Freak was talking about.

I think Joe's answer is definitive, and will remain so until a code-watcher spots something new.

I expect they are just seeing pre Match 8th ones like everyone else may still have.

---

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

Nor are mine, or anyone else's, I expect. I've no idea what XP_Freak was talking about.

I think Joe's answer is definitive, and will remain so until a code-watcher spots something new.

Sorry, I was wrong.
I saw a few pending credits, but did not realise that they were not new.

---

Goodbye Seti Classic