| 00:00.18 | *** join/#storm _mup_ (~mup@li37-90.members.linode.com) |
| 01:51.11 | *** join/#storm munderwo-work (~markl@203-206-133-216.perm.iinet.net.au) |
| 01:51.43 | munderwo-work | Hi all. does anybody know if the debug tracer is threadsafe? |
| 01:52.00 | munderwo-work | ie, will it only record things happening in a single store? |
| 01:52.14 | munderwo-work | or if there are multiple stores open will it debug those as well? |
| 03:24.31 | *** join/#storm niemeyer_ (~niemeyer@200-203-57-221.pltce701.dsl.brasiltelecom.net.br) |
| 03:24.31 | *** mode/#storm [+o niemeyer_] by ChanServ |
| 06:03.57 | *** join/#storm stub (~stub@canonical/launchpad/stub) |
| 06:06.59 | *** join/#storm fcorrea_ (~fcorrea@187.3.129.150) |
| 06:28.04 | jkakar | munderwo-work: It'll record things in the current thread. |
| 06:28.45 | jkakar | munderwo-work: Also, you probably know this, but you should typically only be using a single Store per thread. |
| 06:28.50 | jkakar | Okay, /me => bed |
| 06:29.21 | munderwo-work | jkakar: yeah, im pretty sure im doing that. I create a new store for each request in my wsgi app.. so im pretty sure thats ok |
| 06:29.27 | munderwo-work | jkakar |
| 06:29.30 | munderwo-work | cheers! |
| 06:29.32 | jkakar | munderwo-work: Yeah, that sounds good. |
| 06:29.48 | jkakar | munderwo-work: The debug tracer is really useful, as I guess you know... but the tracing system in general is very good. |
| 06:30.17 | jkakar | munderwo-work: For example, we use it to record all the statements that happen in a given web request and persist them to a log file... each request is given a unique ID, so when things go wrong, or are slow, we can go back and see exactly what happened. |
| 06:30.35 | munderwo-work | jkakar: yeah we are using it for an auditing system |
| 06:30.39 | jkakar | It's really useful and not very hard to put in place. If you look at the code in storm.tracer you can easily figure out how to wire up that kind of thing. |
| 06:31.14 | jkakar | munderwo-work: Another useful thing, which is semi-related, is the TimeoutTracer. You can use it to make sure a request is killed if it spends more than N amount of time in the database. |
| 06:31.31 | jkakar | It's useful if you accidentally end up running a query that will take forever and use a lot of resources. |
| 06:36.58 | lifeless | jkakar: orly :) |
| 06:37.13 | jkakar | lifeless: You might find it useful... you should check it out! |
| 06:37.18 | jkakar | ;b |
| 06:37.27 | jkakar | Okay, bed for reals. :) |
| 06:37.53 | munderwo-work | jkakar: cheers! |
| 09:11.33 | *** part/#storm munderwo-work (~markl@203-206-133-216.perm.iinet.net.au) |
| 11:36.10 | *** join/#storm m4v (~znc@unaffiliated/m4v) |
| 11:44.29 | *** join/#storm stub (~stub@canonical/launchpad/stub) |
| 13:26.12 | *** join/#storm salgado` (~salgado@91.189.93.65) |
| 13:27.38 | *** join/#storm niemeyer (~niemeyer@200-203-57-221.pltce701.dsl.brasiltelecom.net.br) |
| 13:27.39 | *** mode/#storm [+o niemeyer] by ChanServ |
| 15:05.07 | *** join/#storm m4v (~znc@unaffiliated/m4v) |
| 15:21.36 | *** join/#storm m4v (~znc@unaffiliated/m4v) |
| 17:05.52 | *** join/#storm m4v (~znc@unaffiliated/m4v) |
| 17:11.20 | *** join/#storm wgrant (~wgrant@ubuntu/member/wgrant) |
| 17:46.45 | *** join/#storm webm0nk3y (~quassel@adsl-74-235-230-45.mco.bellsouth.net) |