Sorry, a progress indicator for O2AS has been on my todo-list for a few weeks, but currently we are all busy with issues related to the GDPR (deadline May 25). Will do ASAP.
Getting "no work sent" for a bunch of machines. Server status page shows (last time I looked) 239 ready to send. I am guessing that the low number of ready to send and large number of hosts are causing the issue. No I'm not returning errors and the scheduler requests look fine on the server side. Link to one here. This was a core i7 than got 6 work units but can't seem to get another 2 to keep all cores busy.
@Bernd any chance we can pump up the ready to send or is it going as fast as it can?
I got a work unit (h1_0441.10_O2C02Cl1In0__O2AS20-500_441.25Hz_1156_2) validated correctly (and earned a whopping 1000 credits for it) even though my Linux machine only spent 2 seconds on it. The Windows machine invested some 78000s on the same.
I got a work unit (h1_0441.10_O2C02Cl1In0__O2AS20-500_441.25Hz_1156_2) validated correctly (and earned a whopping 1000 credits for it) even though my Linux machine only spent 2 seconds on it. The Windows machine invested some 78000s on the same.
Follow the link to the result that you posted and scroll down to the stderr output section.
The first time stamp in the log for when the task started is: 2018-06-30 03:41:23.0424
And the last time stamp for when the task finished is: 2018-06-30 12:34:28.1330
Rough math puts it at close to 9 hours of wall clock time to complete the task or ~32400s.
Also note that the tasks seems to have been restarted after it was finished as there's two lines reading:
2018-06-30 12:31:06.3042 (5399) [normal]: done. calling boinc_finish(0).
12:31:06 (5399): called boinc_finish
and
2018-06-30 12:34:28.1330 (12415) [normal]: done. calling boinc_finish(0).
12:34:28 (12415): called boinc_finish
I suspect that Boinc lost track of the run time for the task when the restart happened.
I suspect that Boinc lost track of the run time for the task when the restart happened.
I've seen this happen before as well. Not so much with a normal restart but more from the restart after some sort of crash or lockup. When the task restarts from the last checkpoint, there doesn't seem to be a saved time so if the task is near the end anyway, the crunch time returned is ridiculously small. The first couple of times I noticed something like this, I did check that the tasks validated and when they did, I figured it was pretty much as you explained.
Sorry, a progress indicator
)
Sorry, a progress indicator for O2AS has been on my todo-list for a few weeks, but currently we are all busy with issues related to the GDPR (deadline May 25). Will do ASAP.
BM
No, it is not. There is no
)
No, it is not.
There is no search progress indicator as we have for
FGRPB1G
BRP4
or
FGRP5
Bernd Machenschalk
)
Groan. My sympathies. Onwards & upwards ...
Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
Is not there.
)
a
Getting "no work sent" for a
)
Getting "no work sent" for a bunch of machines. Server status page shows (last time I looked) 239 ready to send. I am guessing that the low number of ready to send and large number of hosts are causing the issue. No I'm not returning errors and the scheduler requests look fine on the server side. Link to one here. This was a core i7 than got 6 work units but can't seem to get another 2 to keep all cores busy.
@Bernd any chance we can pump up the ready to send or is it going as fast as it can?
BOINC blog
Hello, I got a work unit
)
Hello,
I got a work unit (h1_0441.10_O2C02Cl1In0__O2AS20-500_441.25Hz_1156_2) validated correctly (and earned a whopping 1000 credits for it) even though my Linux machine only spent 2 seconds on it. The Windows machine invested some 78000s on the same.
https://einsteinathome.org/de/workunit/357250521
This somehow looks suspicious.
Cheers,
Steffen
steffen_moeller
)
Follow the link to the result that you posted and scroll down to the stderr output section.
The first time stamp in the log for when the task started is: 2018-06-30 03:41:23.0424
And the last time stamp for when the task finished is: 2018-06-30 12:34:28.1330
Rough math puts it at close to 9 hours of wall clock time to complete the task or ~32400s.
Also note that the tasks seems to have been restarted after it was finished as there's two lines reading:
I suspect that Boinc lost track of the run time for the task when the restart happened.
Holmis wrote:I suspect that
)
I've seen this happen before as well. Not so much with a normal restart but more from the restart after some sort of crash or lockup. When the task restarts from the last checkpoint, there doesn't seem to be a saved time so if the task is near the end anyway, the crunch time returned is ridiculously small. The first couple of times I noticed something like this, I did check that the tasks validated and when they did, I figured it was pretty much as you explained.
Cheers,
Gary.
Today I recieve 2 tasks which
)
Today I recieve 2 tasks which claim to already been validated. I instinct is to abort them and not waste the electricity. Any comments?
https://einsteinathome.org/workunit/363530444
363530454
I've had those....you'll
)
I've had those....you'll still get credits.
e.g.: https://einsteinathome.org/cs/workunit/362498723