Temporarily failed upload of:

* 05my04aa.7961.21250.392322.115_1_0

BOINC keeps backing off the transfer of this file. Everything else was working fine/has been working fine. Should I be worried?

Is it just because the server is busy? I remember reading something about if it backs up to 1 hour, it will abandon the upload and start a new project. Will I get credit for this project? IOW, will this project continuously try to upload after 1 hour?

Lex

Server is down I think....

Same here!!

Yea,
server must be down

completed work now backing up in transfers

How hard would it be to put code in that shows clearly when and if the server is down? I mean, if the upload keeps getting rejected, is there a reason? What exactly is the reason? Especially if the reason was initiated by the admins? If it's a program lockup or connection failure, at least provide a safeguard for repeat tries for upload. At least give us that.

We shouldn't have to lose our WU, should we?

Lex

> How hard would it be to put code in that shows clearly when and if the server
> is down? I mean, if the upload keeps getting rejected, is there a reason?
> What exactly is the reason? Especially if the reason was initiated by the
> admins? If it's a program lockup or connection failure, at least provide a
> safeguard for repeat tries for upload. At least give us that.
>
> We shouldn't have to lose our WU, should we?
>
> Lex
>
>

If the server is down which it seems to be BOINC will keep trying to upload it untill it succeeds. So it will upload as soon as the server comes back on line. You will get your credit sooner or later.

i dont think it dumps the temp failed at an hour because ive had them get to three before
~boinc its not the credit or satisfaction its the screensavers~
S@h Berkeley's Staff Friends Club © members

>If it's a program lockup or connection failure, at least provide a
> safeguard for repeat tries for upload. At least give us that.
>
> We shouldn't have to lose our WU, should we?
>
> Lex
>
>
'Repeat failed uploads' is exactly what the program will do.

You will not lose any work.

When it comes back 'OnLine' it will upload and 'Report' the completed work units. You can also look under the 'Transfers Tab' and it will show you the count down time for each wu that it is going to try to upload.
----------



Join the Overclockers.com SETI Team!

Failed upload 14:02:07 local Beijing time.
19:36:06....no change...server still down.

Lex,

> How hard would it be to put code in that shows clearly when and if the server
> is down? I mean, if the upload keeps getting rejected, is there a reason?
> What exactly is the reason? Especially if the reason was initiated by the
> admins? If it's a program lockup or connection failure, at least provide a
> safeguard for repeat tries for upload. At least give us that.

The problem sounds simple but it is not. For example, the failure may be a hardware failure or a software failure, or both. In the case of many software failures, the hardware is up, but has no way to send the distress message because the software failed. In the case of hardware failures, in theory, the software stops and that is detectable using a "heartbeat" system. But, these don't work reliably at this stage of our capability.

> We shouldn't have to lose our WU, should we?

You won't. The exponential back-off is doing what it should, it is tryiing, then pausing waiting for the servers to start to respond. As others have said, at that time the system will succeed and the work will go up. If you are running low, well, that is the reason that we are going to a multi-project capable system.

For example, I have 5 projects "running" on BOINC right now...

SETI@Home
Predictor@Home (no new accounts - temporarily off line while they migrate to 4.0)
LHC@Home (no new accounts - may open back up at the end of september)
Einstein@Home (or Pirates@Home - no work at this time)
cp.net (a.k.a. Climate Prediction.net - workunits ~500 hours)

So, while SETI is stalled, I have work from LHC and cp.net in the loop. And I have not actually run out of SETI work yet on most systems (though I am getting very low on most).

If you are not on any of the other projects I would suggest that you at least also join cp.net on BOINC and that will keep you busy while SETI is getting back on the air.
<p>
Click Me!

@Paul Buck

Joining CPDN is only an option if you don't have slower machines (above 800Mhz I think).

> @Paul Buck
>
> Joining CPDN is only an option if you don't have slower machines (above 800Mhz
> I think).

You are right ... shows how much I need to read ...

Yes, 800 MHz is their minimum recommendation. I am also hoping that both Predictor and LHC re-open for new accounts, though it is most likely that we would see this in LHC@Home than Predictor in the near term.

I was lucky enough to get an account in all 5 so far, so, it is keeping my machines busy ...

<p>
Click Me!