[Bug 690640] Re: dh_installinit upstart support start a job in postinst script regardless of previous status

2011-01-26 Thread Clint Byrum
I like that idea quite a bit for natty and later, thats probably the right way to go, especially since we'll also be getting override files at some point. I'll submit it to upstart as a feature request. I doubt we'll backport manual to maverick and lucid's upstart, so we're still left with the

Re: [Bug 690640] Re: dh_installinit upstart support start a job in postinst script regardless of previous status

2011-01-26 Thread Steve Langasek
On Wed, Jan 26, 2011 at 09:28:00AM -, Clint Byrum wrote: I like that idea quite a bit for natty and later, thats probably the right way to go, especially since we'll also be getting override files at some point. I'll submit it to upstart as a feature request. Ok. I doubt we'll backport

Re: [Bug 690640] Re: dh_installinit upstart support start a job in postinst script regardless of previous status

2011-01-25 Thread Steve Langasek
On Tue, Jan 25, 2011 at 03:00:32PM -, Alessandro Bono wrote: Is it possible to backport this fix to Lucid? I tried your patch on lucid's debhelper and packages compiled with this patched version has correct scripts I think a backport to lucid is of questionable benefit, unfortunately,

[Bug 690640] Re: dh_installinit upstart support start a job in postinst script regardless of previous status

2011-01-25 Thread Clint Byrum
By my estimate, there are a bit over 50 pending, unassigned SRU bugs for the server team in lucid. Many of these are small but the time just to get them into the pipeline has been scarce and we are under-staffed as it is. Also the SRU team isn't exactly bursting at the seams. So while I do think

Re: [Bug 690640] Re: dh_installinit upstart support start a job in postinst script regardless of previous status

2011-01-25 Thread Steve Langasek
On Tue, Jan 25, 2011 at 08:04:15PM -, Clint Byrum wrote: With upstart jobs, its not really possible to know if a job should be running by looking at the filesystem. So this seems to be missing a fix that would, as Alessandro said, query and save the status when the stop action is run

[Bug 690640] Re: dh_installinit upstart support start a job in postinst script regardless of previous status

2011-01-25 Thread Clint Byrum
Alright, I see where we do need this information encoded. The current recommendation for natty is to add the word 'manual' to the job configuration file to disable it. For 10.10 and earlier, commenting out the start on line is the preferred method, or moving the file somewhere else. What if we

[Bug 690640] Re: dh_installinit upstart support start a job in postinst script regardless of previous status

2011-01-25 Thread Alessandro Bono
Lucid backport is a good thing to do because also if we don't rebuild all affected package now at first update we resolve a bug for free, lucid is a LTS and probability of update package in future is high -- You received this bug notification because you are a member of Ubuntu Server Team, which

Re: [Bug 690640] Re: dh_installinit upstart support start a job in postinst script regardless of previous status

2011-01-25 Thread Steve Langasek
On Tue, Jan 25, 2011 at 09:15:25PM -, Clint Byrum wrote: Alright, I see where we do need this information encoded. The current recommendation for natty is to add the word 'manual' to the job configuration file to disable it. For 10.10 and earlier, commenting out the start on line is the