I found hundreds of results with credit 0.
One of them is http://setiweb.ssl.berkeley.edu/result.php?resultid=158348295
In News is : "We will be extending the deadline for returning results so that the troubles with the result upload handler will not result in lost credit.".
Where is problem?
Thanks
Libor

---

The reason maybe because that example was validated for the other three hosts before the problems 28 Nov 2005.

---

His deadline for that unit would have been the 10th. I think he should have been granted credit on that one. I suspect the transitioner/deleter/purger did their work prior to the outage. I assume they artificially extended deadline by turn off the validators, but am unsure. Seems to be a timing issue of when they turned what off.

The WU would have been validated when the third result came in on the 28th, and the credit awarded then. Is it even the validator that would pick up the fourth result when it came in and decide if it was within deadline? Or is it another process? I was thinking it was the validator, making another pass, but if that ran on the 28th and is NOT running now, but something has "looked at" that result on the 13th... I'm lost.

There's definitely a disconnect between the news statement "will not result in lost credit" and the current state of this particular result. I'm sure it's fixable, but someone will have to get the attention of whoever is needed to fix it...

Tony,

For the fourth result in on a unit that has been validated, do you know wether the fourth unit gets to the validate process or is it handled in the 'transitioner'.

There seems to be no clear explanation that I can find.

---

I don't know, something seems hinckie with that wu. I don't know about the rest he claims. I think this might turn into a bigger issue by the end of the day.

If he would "unhide" his puters, I could take a closer look. There is NO sensitive data display when you "unhide" your puter. Click on my username (in the left column), select View puters, and see for yourself. What you see when you do that to your puters is different from what I see when I do it.

---

I don't know, something seems hinckie with that wu. I don't know about the rest he claims. I think this might turn into a bigger issue by the end of the day.

when looking at Team Seti's status graphs it seems the file-deleters had a good run last thursday.
there might not be anything left to validate results with a deadline in the first part of the outage against.

I already posted a past-deadline-result yesterday in another thread.

yer, my laptop finaly got it's 1 upload through after four and a half (work)days of trying.
I just hope the extended deadlines also include results that should have been returned the 9th as it is showing in red with a "no reply".

The same thing happened to me, too:

http://setiathome.berkeley.edu/workunit.php?wuid=37966816

This workunit, which was completed in time, eventually timed out because of the server problem. I'm convinced this is a general problem.
Does anybody know if there is an official statement by S@H about this?

---

The same thing happened to me, too:

http://setiathome.berkeley.edu/workunit.php?wuid=37966816

This workunit, which was completed in time, eventually timed out because of the server problem. I'm convinced this is a general problem.
Does anybody know if there is an official statement by S@H about this?

Probably not from Berkeley, its about Four in the morning there.

---

Oi, They should start pouring in now.

Thomas, they aren't even out of bed yet to notice this, let alone make any kind of statement. I'd recommend waiting to see how big this gets, and then what berkeley has to say.

There is nothing to be done now anyway, sit back, relax, watch with "keen" interest is about all you can do.

Oi, They should start pouring in now.

Thomas, they aren't even out of bed yet to notice this, let alone make any kind of statement. I'd recommend waiting to see how big this gets, and then what berkeley has to say.

There is nothing to be done now anyway, sit back, relax, watch with "keen" interest is about all you can do.

Sounds like good advice to me.
There must be a terrible mess to clean up on their servers anyway.

---

Hi,

I have four results, which got "Result was reported too late to validate":
WU #38120243
WU #38120241
WU #38120230
WU #38120184

As far as I remember from the last time, they got this status immediatly after the report. So, whatever marked these results as "too late", it must have happended sometime between the deadline and the report.

Maybe worth to note: The status of one of my results, which had the same deadline like the others, is still "pending". The only difference, I can figure out, is that the particular WU had "client errors" and my result was a re-issued result:
WU #37691853

(Please note: I do not complain ... ;-) I only give some examples. Maybe they are useful for somebody, who can figure out, what exactly happened.)

Regards,
Carsten

Just a personal observation, but during the troubles we went through a period when uploads were blocked and downloads allowed. If in this period the validators etc were still running we could have units timed out, incorrectly, and replacments reissued. This could lead to some interesting times.

---

All of my WU's dated to end the 10th and 11th have ended with 0 credit

Servers must be a huge mess

---

Same here. I have many results "Too late to validate". Will the promise of extending deadlines be executed at a later date? Perhaps someone with inside knowledge could answer this.

Mine are all still pending.
But they are pending with the original date!
I think the extended time period is still not applied yet!

Greetings from Berlin, Germany

Uwe

---

This happened the last outage. I had many at 0 for a few days. They then ran a special program to go back and verify the overdue ones and I got credit. Be patient. Allow the GURUs to do their work.

---

My movie https://vimeo.com/manage/videos/502242

I've got several WU's in a similar state. I could be wrong, but it looks like if credit had been calculated before the outage (and I was turning in the 4th result), the too-late-to-validate flag got set; workunits that didn't have credit calculated yet are marked "pending" and will presumably be credited in the normal course of events.

Having counseled patience in the past, I'm willing to take my own advice, but will be, um, watching with interest.

Happy crunching, everyone!

---

If you think education is expensive, try ignorance.

I have four results, which got "Result was reported too late to validate":
WU #38120243
WU #38120241
WU #38120230
WU #38120184

The transitioner will make this decision. But, the problem is one of timing. Worse luck is that it is not likely that this can be reversed if the Canonical Result is gone (assimilated).

Also bad is the fact that you blew the reporting deadline anyway.

But, the ones I looked at, the others reported them early before the troubles. The validated and were waiting for the expiration of the deadline. That happened before you got your results back.

If this is a big issue for you, send a note to Dr. Anderson.

This is one of the potential drawbacks for not reporting every day ... :(

We are trying to get a change through that will break the link between the queue size and the connect interval. No one has come up with an iron-clad design yet that passes muster (that I am aware of) and I am not sure what priority UCB would put on it ... or if there is another volunteer developer panting at the chance to make the changes ...

---