| __TJ__ | does upstart facilitate in some kind of statics ? if we have upstart automatically respawn applications when any has crashed, we would like to know how many times this has happened..... | 08:49 |
|---|---|---|
| jhunt_ | __TJ__: no, no statistics at present. The 3 main options you have currently: (1) run upstart with log-priority of "debug" and grep the system log for restart messages, (2) log your own statistics by having a script section write a log each time the application starts, or (3) create a new job which looks for the main job starting and writes some statistics. | 09:21 |
| jhunt_ | I'll add details on this to the Upstart Cookbook... | 09:21 |
| __TJ__ | jhunt_: thanks.... was just now thinking about some script or binary that will be ran by the upstart script itself by the pre-start section.... | 09:22 |
| __TJ__ | the only thing needed then will be the pre-start command being run also when respawn is done | 09:23 |
| __TJ__ | but i assume thats the case | 09:23 |
| __TJ__ | jhunt: using the pre start for the statistics is also a good approach ? | 09:43 |
| codebeaker | is there any recommended way to show upstart statuses inside of a web app? I'd like to offer my users the ability to emit signals and see what's running according to the initctl daemon | 13:08 |
| jhunt_ | sounds potentially dangerous to me, but if you really want to do this, but d-bus gives you a consistent api. Examples using dbus-send: http://upstart.ubuntu.com/cookbook/#controlling-upstart-using-d-bus | 13:12 |
| codebeaker | thanks jhunt_ I think it's a bad idea too :) but I've been tasked with weighing it up, and seeing it it's at least viable | 13:12 |
| codebeaker | thanks jhunt_ - that looks good - upstart docs are getting better all the time | 13:13 |
| jhunt_ | codebeaker: np. | 13:13 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!