Long running work unit

[Trotador]

Btw, I've just realised that I've passed the deadline for this one

http://www.rnaworld.de/rnaworld/result. ... d=11421786

It is now almost in 500 hours, would it be possible to extend the deadline? No one has finished it yet

Thanks for looking at it.

regards,

[mxplm]

Deadline extended until 20th of april. happy crunching ;)

EDIT: RNA Page says:

Status:Timed out - no response
Server state: Over
Outcome: No reply

@yoyo: Does this mean the server already noticed that the Result is too late?

[yoyo]

Deadline extended until 20th of april. happy crunching

EDIT: RNA Page says:

Status:Timed out - no response
Server state: Over
Outcome: No reply

@yoyo: Does this mean the server already noticed that the Result is too late?

Yes, but you can also reset the states of this result:
(server_state, outcome, client_state, validate_state, delete_state) = (4, 0, 0, 0, 0)
yoyo

[mxplm]

...of course you can... could have thought of that myself ^_^

ok, done.

[yoyo]

Did you change it?
yoyo

[MReed]

@Yoyo:

ok, done.

[mxplm]

@Yoyo:

ok, done.

:wink:

right ^_^

[zombie67]

cms_GA[e30-50MB_Lin64f]_Oryza-sativa-Japonica-Group_AP008209.lin.EMBL_RF00028_Intron_gpI_1293996458_0
WU. Created on January 2nd.
task. Sent to my machine on January 18th.
Hours spent running so far: 424

Completed after 876 hours, waiting for a wingman. Only two still in process.

"Too many error results"

So now what?

Edit: I know tasks don't live in the db forever. So just for the record, my machine claimed 20,676.24.

[Uwe Sänger Herzke]

OK, you did it in 670h, congratz for that , I'm now at 417h, if I compare our benchmarks, I will crunch for another ~150-200h, that's less than 10 days, yeah!

Go for it, I need a wingmen that finish the WU!!

Btw at what clock are you running your processor?

I'll do my very best

The machine is running at 3.2GHz

After 657h the WU finally finished, but got invalidated.
It got as well some strange numbers allocated, which I can't see how they were generated:
Run time: 2,365,527.74 (looks fine)
CPU-time: 46,116,860,814,263,600.00 (very odd)
Credit claim: 342,629,280,138,119.00 (Can I have it? )

And somehow it got invalidated, perhaps because of that strange numbers.

[Michael H.W. Weber]

Something strange happened with my Introns, too. I had two of these running with runtimes of around 750 hrs. The third was started about 48 hrs ago. Now, the actual CPU time of all three of these WUs is indicated as 35 hrs in the taskmanager. So, I guess these got a problem when the third WU reached its peak memory demand (which usually happens several hours after start of WU computation). The machine (ID 169, a Phenom II 955 BE) has 8 GB of physical RAM and and a swap file of 24 GB. Apparently this was insufficient and reminds me to never allow more than two of the Intron WUs to be computed at the same time on these types of machines.

Michael.

[yoyo]

cms_GA[e30-50MB_Lin64f]_Oryza-sativa-Japonica-Group_AP008209.lin.EMBL_RF00028_Intron_gpI_1293996458_0
WU. Created on January 2nd.
task. Sent to my machine on January 18th.
Hours spent running so far: 424

Completed after 876 hours, waiting for a wingman. Only two still in process.

"Too many error results"

So now what?

Edit: I know tasks don't live in the db forever. So just for the record, my machine claimed 20,676.24.

Yes, Boinc server aborts a WU if there are to many error results.
But anyway, I granted you some credits for this result.
yoyo

[yoyo]

After 657h the WU finally finished, but got invalidated.
It got as well some strange numbers allocated, which I can't see how they were generated:
Run time: 2,365,527.74 (looks fine)
CPU-time: 46,116,860,814,263,600.00 (very odd)
Credit claim: 342,629,280,138,119.00 (Can I have it? )

And somehow it got invalidated, perhaps because of that strange numbers.

Which workunit/result was it?
yoyo