Well, I, for one, have calmed down.
I have to resign myself to the fact that since I have chosen to crunch Seti, and Seti only, I must accept the downtime of the project gracefully.

And there will be some times when my rigs cannot get enough work on hand to keep them running during extended or sequential project failures.

Such is what I have chosen to do.

Welcome to life in Setiland.

---

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

Well, I, for one, have calmed down.
I have to resign myself to the fact that since I have chosen to crunch Seti, and Seti only, I must accept the downtime of the project gracefully.

And there will be some times when my rigs cannot get enough work on hand to keep them running during extended or sequential project failures.

Such is what I have chosen to do.

Welcome to life in Setiland.

Mark, that is the most marvellous post I think I have ever seen you make. I always knew it was worth believing in you.

Take care now.

Chris S.

The calm is nice to see and appreciated.

---

glad to see u back

funny msg:

21.06.2010 08:19:04 [error] Error reported by file upload server: Server is out of disk space

its no wonder with over 1 million unvalidated WU

---

- Performance is not a simple linear function of the number of CPUs you throw at the problem. -

funny msg:

21.06.2010 08:19:04 [error] Error reported by file upload server: Server is out of disk space

its no wonder with over 1 million unvalidated WU

Got this message as well today with an upload transient error, will just sit and wait until it has been fixed

---

funny msg:

21.06.2010 08:19:04 [error] Error reported by file upload server: Server is out of disk space

its no wonder with over 1 million unvalidated WU

Got this message as well today with an upload transient error, will just sit and wait until it has been fixed

I turned off network access in Boinc earlier when I heard You all talking about It, I'm sure It'll get fixed on Monday or Tuesday, So We wait.

---

The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's

Had to happen sooner or later.
If the validators are switched off then the results are not been transferred to the science database. So no space is being emptied to make room for further uploads.

Actually it would help much more if the assimilators would do their job. As long as they are not working, validating the results don't help to free up some disk space.

---

Now we have really hit the end of the world. Not only are the project developers playfully screwing around with the boinc structure, apparently affecting not just SETI but many other projects per some other posts around here, but Scarecrow is off-line due to thunderstorms in Nebraska, or so it would seem. What's next? A rain of fire and the face of evil leading us to the End of Days?

Now we have really hit the end of the world. Not only are the project developers playfully screwing around with the boinc structure, apparently affecting not just SETI but many other projects per some other posts around here, but Scarecrow is off-line due to thunderstorms in Nebraska, or so it would seem. What's next? A rain of fire and the face of evil leading us to the End of Days?

Oh no...(Putting on tin hat)...It is the end.

---

Official Abuser of Boinc Buttons...
And no good credit hound!

9 million result files averaging 25KB would be about 225GB of disk space. I don't know how big bruno's fiber channel array is, but would have guessed at least 2TB.

                                                               Joe

"Oh no...(Putting on tin hat)...It is the end."

Not again!!!

;-)

9 million result files averaging 25KB would be about 225GB of disk space. I don't know how big bruno's fiber channel array is, but would have guessed at least 2TB.

                                                               Joe

The problem is more likely to be the (8,672,596 - 1,058,977) = 7,613,619 workunit data files @ 367 KB - that's 2.6 TB that can't be deleted until the validator decides whether they need to be resent or not.

Not to mention the 68,714 Astropulse results in the same state - that's another half a terabyte for the 8 MB data files.

Errors in both directions on those figures, of course - there will be some triple returns still inconclusive, and some datafiles with no returns at all (both tasks srill out in the field), but the order of magnitude should be right if my calculator is working properly...

9 million result files averaging 25KB would be about 225GB of disk space. I don't know how big bruno's fiber channel array is, but would have guessed at least 2TB.

                                                               Joe

But how much of the 2TB is available for the validator? Bruno does some other stuff, and maybe the disk is full of other files, like saved games and photos from last summer's vacation (based on a quick survey of what's filled up my disc).

But seriously, how long will it take to clear the validator back log? Maybe the Tuesday shutdown will come a day early this week.

---

9 million result files averaging 25KB would be about 225GB of disk space. I don't know how big bruno's fiber channel array is, but would have guessed at least 2TB.

                                                               Joe

The problem is more likely to be the (8,672,596 - 1,058,977) = 7,613,619 workunit data files @ 367 KB - that's 2.6 TB that can't be deleted until the validator decides whether they need to be resent or not.

Not to mention the 68,714 Astropulse results in the same state - that's another half a terabyte for the 8 MB data files.

Errors in both directions on those figures, of course - there will be some triple returns still inconclusive, and some datafiles with no returns at all (both tasks srill out in the field), but the order of magnitude should be right if my calculator is working properly...

I agree, it's whatever else is being stored in bruno's array that is likely causing the error. We can only guess how they've got storage mapped.

Here's a very unlikely thought; awhile back IIRC Eric posted about a difficulty with the sah_validate processes running briefly each time they were started, then falling over. Suppose that same difficulty has returned but they have a cron job set to restart them periodically and every time they crash they have consumed some of bruno's disk array space...

                                                                Joe

...
Here's a very unlikely thought; awhile back IIRC Eric posted about a difficulty with the sah_validate processes running briefly each time they were started, then falling over. Suppose that same difficulty has returned but they have a cron job set to restart them periodically and every time they crash they have consumed some of bruno's disk array space...

                                                                Joe

I remember that. I believe it was just for AP though.

---

Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)