Well, related or unrelated, we are seeing performance problems in our replica database since the power outage, e.g. the server status page is refreshing way slower than it used to do. We are investigating this at the moment, and I would not be surprised to find out that the pending credit page is also affected by this. Anyway bear with us, we hope to resolve this soon.
Cheers
HB
Another aditional info:
When my pendings WU downs bellow 2000 then the page restarts to work normaly.
Anyway, the pending credit page in its current form (this one: http://einstein.phys.uwm.edu/pending.php) probably is not too useful on E@H anyway. It does not brake down the results in pages, this might explain its slowness/not working at all for long result sets.
If you want to check on the pending tasks across all searches and all your computers, there's also the overall task list that can be used (link http://einstein.phys.uwm.edu/results.php?userid= ) under "Your Account -> Tasks", then click on "Pending". This gives you a paged (20 results per page) view, and this should still work even for long lists.
The only additional thing that the "Pending credit" page does is to compute a sum of the claimed credit for all pending tasks, but as E@H assigns a fixed credit per task, this number is completely useless on E@H anyway.
If you want to check on the pending tasks across all searches and all your computers, there's also the overall task list that can be used (link http://einstein.phys.uwm.edu/results.php?userid= ) under "Your Account -> Tasks", then click on "Pending". This gives you a paged (20 results per page) view, and this should still work even for long lists.
Cheers
HB
Seti has similar, with one addition, after each of the filters there is a total count for each filter, a very useful guide if you are working with larger numbers of tasks.
If you want to check on the pending tasks across all searches and all your computers, there's also the overall task list that can be used (link http://einstein.phys.uwm.edu/results.php?userid= ) under "Your Account -> Tasks", then click on "Pending". This gives you a paged (20 results per page) view, and this should still work even for long lists.
Cheers
HB
Seti has similar, with one addition, after each of the filters there is a total count for each filter, a very useful guide if you are working with larger numbers of tasks.
Kevin
that feature often saves going in to the lists at all. More often than not just the count is enough. (2 errors for 986 valid? close enough... )
Seti has similar, with one addition, after each of the filters there is a total count for each filter, a very useful guide if you are working with larger numbers of tasks.
Kevin
that feature often saves going in to the lists at all. More often than not just the count is enough. (2 errors for 986 valid? close enough... )
Those filters counts are a nifty feature, and I suspect the computational cost of supplying them may be modest, once implemented.
As a user, I'd love to see a feature of implementing such counts get onto the Einstein implementation priority list.
Seti has similar, with one addition, after each of the filters there is a total count for each filter, a very useful guide if you are working with larger numbers of tasks.
Kevin
that feature often saves going in to the lists at all. More often than not just the count is enough. (2 errors for 986 valid? close enough... )
Those filters counts are a nifty feature, and I suspect the computational cost of supplying them may be modest, once implemented.
As a user, I'd love to see a feature of implementing such counts get onto the Einstein implementation priority list.
+1
David
Miserable old git
Patiently waiting for the asteroid with my name on it.
As a user, I'd love to see a feature of implementing such counts get onto the Einstein implementation priority list.
The problem is that the DB replica is unusually slow already, I'm quite reluctant to implement anything that would put even more load on it before this problem is solved.
Well, related or unrelated, we are seeing performance problems in our replica database since the power outage, e.g. the server status page is refreshing way slower than it used to do. We are investigating this at the moment, and I would not be surprised to find out that the pending credit page is also affected by this. Anyway bear with us, we hope to resolve this soon.
Cheers
HB
Another aditional info:
When my pendings WU downs bellow 2000 then the page restarts to work normaly.
Another clue...
My pendings rises above 2000... and the page stops to work again...
RE: Working now.... I
)
I realy hate that part of the day... going too...
RE: Hi Well, related or
)
Another aditional info:
When my pendings WU downs bellow 2000 then the page restarts to work normaly.
Hmmm...I see. Anyway, the
)
Hmmm...I see.
Anyway, the pending credit page in its current form (this one: http://einstein.phys.uwm.edu/pending.php) probably is not too useful on E@H anyway. It does not brake down the results in pages, this might explain its slowness/not working at all for long result sets.
If you want to check on the pending tasks across all searches and all your computers, there's also the overall task list that can be used (link http://einstein.phys.uwm.edu/results.php?userid= ) under "Your Account -> Tasks", then click on "Pending". This gives you a paged (20 results per page) view, and this should still work even for long lists.
The only additional thing that the "Pending credit" page does is to compute a sum of the claimed credit for all pending tasks, but as E@H assigns a fixed credit per task, this number is completely useless on E@H anyway.
Cheers
HB
RE: . If you want to
)
Seti has similar, with one addition, after each of the filters there is a total count for each filter, a very useful guide if you are working with larger numbers of tasks.
Kevin
RE: RE: . If you want
)
that feature often saves going in to the lists at all. More often than not just the count is enough. (2 errors for 986 valid? close enough... )
RE: RE: Seti has similar,
)
Those filters counts are a nifty feature, and I suspect the computational cost of supplying them may be modest, once implemented.
As a user, I'd love to see a feature of implementing such counts get onto the Einstein implementation priority list.
RE: RE: RE: Seti has
)
+1
David
Miserable old git
Patiently waiting for the asteroid with my name on it.
RE: As a user, I'd love to
)
The problem is that the DB replica is unusually slow already, I'm quite reluctant to implement anything that would put even more load on it before this problem is solved.
BM
BM
RE: RE: Hi Well, related
)
Another clue...
My pendings rises above 2000... and the page stops to work again...
It would seem to me a line
)
It would seem to me a line count would be less taxing on resources than listing them out. But.. I am not a programmer.