=== m3thyl is now known as rdn | ||
Tor-Mentor | Hi! im having some wierd problems with my scripts and bash_completion. Service does not autocomplete my scripts but initctl does. Does it depend on the content in the scripts somehow? | 08:24 |
---|---|---|
jodh | Tor-Mentor: /usr/sbin/service is *not* an upstart command - it's a SysV one. However, to make life more familiar, links are created from /etc/init.d/$foo to the corresponding upstart jobs in /etc/init/ to allow you to continue to use that command. | 08:34 |
jodh | Tor-Mentor: hence, if you want the bash completion, please use initctl. | 08:34 |
Tor-Mentor | aahh thanks. That clears that one up! | 08:37 |
tjaalton | do you folks see anything alarming on this job http://pastebin.com/2bZu24yG | 14:35 |
jodh | tjaalton: you don't need the "console output". It's possible your job will never detect plymouth-splash running though I think as your job is probably going to start after plymouth-splash. | 14:48 |
tjaalton | jodh: oh right, that was cruft.. | 14:49 |
jodh | tjaalton: also, your job is going to run that ping in a very tight loop. | 14:49 |
tjaalton | ugh, yes :) | 14:49 |
jodh | tjaalton: are you looking to put this change into raring? | 14:49 |
tjaalton | jodh: if possible, yes | 14:49 |
tjaalton | ok added 'sleep 0.1 after fi | 14:49 |
jodh | tjaalton: ok - if this goes in though and other jobs are changed to make use of the new event, I lot of testing is going to have to happen in the next week :) | 14:50 |
tjaalton | jodh: yeah.. | 14:50 |
tjaalton | only lightdm for now I think | 14:50 |
jodh | tjaalton: ok. Are QA aware they'll need to test this? | 14:52 |
tjaalton | not yet | 14:52 |
tjaalton | slangasek suggested to use 'start on startup or started plymouth-splash' and it worked this time (didn't before for some reason) | 14:53 |
jodh | tjaalton: just spotted another problem: you 'if' and 'while' reference 'STARTED' where it should be '$STARTED'. | 14:53 |
tjaalton | this can go in plymouth in any case | 14:53 |
tjaalton | testing lightdm can be done as an sru | 14:53 |
tjaalton | duh | 14:53 |
tjaalton | good thing lp timed out when adding it to the bug.. | 14:54 |
tjaalton | :) | 14:54 |
tjaalton | it's a monster bug with 100+ dupes | 14:54 |
tjaalton | ok, the upstart job is now good enough, thanks for your help :) | 17:26 |
elmo | jodh: hey is there an ETA for the upstart dbus bridge and/or anything already written? | 17:33 |
jodh | elmo: no concrete eta, although I was hoping to have something for 'S'. No code yet. We have a dconf/dsettings bridge that's almost complete. | 17:37 |
elmo | jodh: ok, cool, just curious, most of my ideas for user upstart jobs would trigger off of dbus | 17:37 |
slangasek | AIUI a dbus bridge is not currently a priority for the desktop team | 17:41 |
slangasek | we'll double-check that with them next month, though | 17:41 |
elmo | FAOD, my curiousity is entirely idle/spare time and not professional :) | 17:44 |
slangasek | elmo: so you're not writing a job that runs 'juju deploy' based on an incoming indicator-message event? ;) | 17:46 |
elmo | yes, that's my plan for juju auto scale | 17:47 |
elmo | pagerduty + ubuntu webapps + indicator-osd + upstart-dbus-bridge + user upstart job -> juju add-unit | 17:47 |
elmo | :-P | 17:47 |
elmo | (oh god, that's so horrible, I want to bleach out my own mind) | 17:48 |
* slangasek giggles | 17:48 | |
elmo | no, FWIW, I mostly want to be able to see things like network changes and screen lock/unlock events, which AFAIK are announced most usefully on DBus | 17:50 |
* slangasek nods | 17:50 | |
xnox | slangasek: that quote on the wiki, got me scared =) | 17:55 |
slangasek | xnox: is your entire life RSS-driven? :) | 17:56 |
xnox | slangasek: ssshhh | 17:57 |
xnox | slangasek: have you seen this yet http://youtu.be/A25VgNZDQ08 that was my reaction when I heard the news about google reader shutdown. | 17:59 |
slangasek | funny thing is, I had just started using Google Reader on my phone two weeks before the announcement | 18:00 |
xnox | slangasek: yeah, I'm waiting for newsblur to scale, their mobile app is good enough. They are scalability issues, since they got effectively DDOSed after google reader announcement. | 18:01 |
* xnox ponders if Blogger and Feedburner are the next to get the cut. Cause clearly people should blog by doing "public posts on G+" | 18:01 | |
slangasek | "It may amaze you gentlement that Orkut is still up and running" -hahaha | 18:03 |
slangasek | xnox: right... except for the days that G+ is failing to add stuff to your stream | 18:04 |
slangasek | (I had a ~16h outage where people I'm following were not having their posts showing up in my stream - right before the google docs outage) | 18:04 |
xnox | ouch. | 18:04 |
xnox | well..... I can't tell the difference on my G+ feed really. Half of it is cats, the other half is people playing ingress. | 18:05 |
slangasek | you should follow more interesting people | 18:06 |
slangasek | like jodh | 18:06 |
slangasek | O right, all his G+ posts are links to his blog which you already have in RSS ;) | 18:07 |
xnox | =)))) | 18:07 |
xnox | i follow people who reshare interesting rss feed posts via google reader instead =) | 18:08 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!