Re: [MTT users] Reporter problems

2008-01-31 Thread Ethan Mallove
It would be useful to know how long the queries were taking, e.g., on date "x" when the database had "n" tuples. (Idea for the stats table?) I was just going by my [possibly unreliable] recollection of top-level summaries taking only a few seconds. -Ethan On Wed, Jan/30/2008 03:14:20PM, Josh Hur

Re: [MTT users] Reporter problems

2008-01-30 Thread Josh Hursey
I'm seeing between 12 and 20 seconds on a fairly idle machine. We can likely do better. I'll dig into it this week[end] and see what I can do. 12 - 20 isn't too bad though considering the amount of data that query is returning. :) On Jan 30, 2008, at 2:52 PM, Ethan Mallove wrote: I don't

Re: [MTT users] Reporter problems

2008-01-30 Thread Ethan Mallove
I don't remember a "past 24 hour" summary taking "24 seconds". Are we seeing a slow down due to an accumulation of results? I thought the week-long table partitions would prevent this type of effect? -Ethan On Wed, Jan/30/2008 11:00:46AM, Josh Hursey wrote: > This maintenance is complete. The re

Re: [MTT users] Reporter problems

2008-01-30 Thread Josh Hursey
This maintenance is complete. The reporter should be operating as normal. There are a few other maintenance items, but I am pushing them to the weekend since it will result in a bit of a slowdown again. Thanks for your patience. Cheers, Josh On Jan 29, 2008, at 9:47 AM, Josh Hursey wrote:

Re: [MTT users] Reporter problems

2008-01-29 Thread Josh Hursey
The reporter should be responding much better now. I tweaked the maintenance scripts so they no longer push nearly as hard on the database. They are still running, but the query you specified seems to run in approx. 15-20 sec. with the current load. -- Josh On Jan 29, 2008, at 8:38 AM, Jos

Re: [MTT users] Reporter problems

2008-01-29 Thread Josh Hursey
For the next 24 - 48 hours this is to be expected. Sorry :( I started some maintenance work last night, and it is taking a bit longer than I expected (due to integrity constraint checking most likely). The maintenance scripts are pushing fairly hard on the database, so I would expect some s