| keesj | I have about the same problem as kysucix. I need to control system scripts | 08:56 |
|---|---|---|
| keesj | and need to do stuff initctl can do | 08:56 |
| keesj | the pain is that libnih and the upstart libs are gpl so I can not even write my own initctl using the nih libs | 08:56 |
| keesj | in ubuntu 8.10 there is not even a libnih | 08:57 |
| Keybuk | why can't you write your own initctl? | 09:13 |
| Keybuk | you can link to gpl libraries from gpl applications just fine | 09:15 |
| keesj | I am not writing a gpl app | 09:21 |
| keesj | making the libs lgpl would really help | 09:21 |
| Keybuk | *shrug* | 09:22 |
| Keybuk | then I have no interest in helping you | 09:22 |
| Keybuk | if you do not want to contribute your source code | 09:22 |
| Keybuk | then don't touch mine | 09:22 |
| Keybuk | the libraries are GPL for a reason | 09:22 |
| keesj | dont' start the lame gpl story. If I am on this channel its because my company tries to use upstart. | 09:24 |
| keesj | and it 's not because I am writing closed code that I can not help people | 09:24 |
| keesj | just like I did no this channel many times | 09:24 |
| Keybuk | you can write as much closed code as you like | 09:25 |
| Keybuk | you just can't link to my code | 09:25 |
| Keybuk | I am happy that inconveniences you | 09:25 |
| Keybuk | I spent a lot of time and effort on my code | 09:26 |
| keesj | Well you should not. | 09:26 |
| Keybuk | and its open for all to use | 09:26 |
| Keybuk | it's frankly insulting that you want to benefit from it commerically without opening your code in return | 09:27 |
| keesj | I spend a lot of time on upstart and having ideas. if you can't value that. or don't value other good your are probably thinking to much of open-source as a software only thing | 09:28 |
| Keybuk | I'm not talking about having ideas, or helping out here | 09:28 |
| Keybuk | that's more than welcome | 09:28 |
| keesj | it about people. I am really shocked that you don't value whay I do | 09:28 |
| Keybuk | I'm only talking about you asking me to lgpl the libraries | 09:28 |
| Keybuk | which is only a software thing | 09:28 |
| Keybuk | my code is open | 09:30 |
| keesj | I am asking you to understand that upstart like any ohter open-source thing need people can companies behind it. and I represent such a company. | 09:30 |
| Keybuk | and free for you to modify, derive and link as you see fit | 09:30 |
| Keybuk | *provided* that you open yours in return | 09:30 |
| Keybuk | if your company isn't willing to open its code, then I don't see that as backing upstart's code development | 09:32 |
| Keybuk | obviously you can help in other ways | 09:32 |
| keesj | that why lgpl exist .I did not have much needs to modify upstart. it works quite well out of the box (the 0.3.* series) | 09:32 |
| Keybuk | but I'm not interested in a company making commercial derivatives or taking private benefit from my code | 09:33 |
| keesj | I can only help and am only interested in upstart because we want to use it. that is WHY is am here. no matter what I do with it. and apparently you have no way of showing that you value that in code :P | 09:34 |
| Keybuk | I only value code I have the same rights to as code you clearly value | 09:35 |
| Keybuk | (or more rights to :p) | 09:35 |
| Keybuk | code that I'm not allowed to see is inherently uninteresting | 09:35 |
| Keybuk | code that improves my work, that I'm not allowed to see, is ... well, insert rude words here | 09:35 |
| keesj | the result of this disc doesn't really matter (almoost flame stuff). | 09:36 |
| Keybuk | I don't see this as a flame | 09:37 |
| keesj | I did not modify your code. but I want to listen to upstart events from my closed system and that is apparenly aksing to much | 09:37 |
| Keybuk | yes, your closed system benefits from, and is improved by, upstart | 09:38 |
| Keybuk | but we, in return, can't benefit from and be improved by your system | 09:38 |
| Keybuk | because it's closed | 09:38 |
| Keybuk | this is unfair | 09:38 |
| keesj | that part is unfair yes. but that is not the whole story | 09:39 |
| Keybuk | as far as the code and licence goes, that is the *entire* story | 09:39 |
| keesj | you one yout hand will get "upstart is used in blabla". Me who at least tried to help in thinkgs about upstart. me helpin people on irc. and somebody else using your system and giving feedback | 09:40 |
| Keybuk | honestly, and I'm sorry if this sounds conceited, but: | 09:41 |
| Keybuk | between Ubuntu, Fedora and RHEL; I have probably 20-25 million users of Upstart | 09:41 |
| Keybuk | that's such a large level, I could never possibly hope to receive all the feedback | 09:41 |
| Keybuk | so additional users are nice, but not something I really seek | 09:42 |
| Keybuk | my goal is to have it adopted by the major Linux distributions | 09:42 |
| Keybuk | and that's really the end of my goal | 09:42 |
| keesj | what I see is 3 user cases , 1 coder and a few developers | 09:43 |
| Keybuk | it's nice to hear about people trying to use it for other things (hi, Cisco!) but they're so far off my radar, it's not something I worry about | 09:43 |
| keesj | it an enrightment | 09:43 |
| Keybuk | yeah | 09:44 |
| Keybuk | that's 2 use cases too many ;) | 09:44 |
| keesj | Well I am in the embedded PND world | 09:45 |
| Keybuk | Post Natal Depression? | 09:45 |
| Keybuk | (that's the only expansion of that tla I know) | 09:45 |
| keesj | Portable navigation devices | 09:46 |
| Keybuk | GPS units? | 09:46 |
| keesj | I wonder what upstart version maemo will be using... | 09:47 |
| keesj | yes GPS | 09:47 |
| Keybuk | keesj: I don't know, they've never even been in touch to say that they would be using it | 09:48 |
| Keybuk | I found out because someone showed me a slide from a presentation | 09:48 |
| Keybuk | I guess bugs or patches will land in due course | 09:48 |
| Keybuk | (or it works perfectly, which I'd be surprised about, because I'm pretty sure our arm build threw a nut) | 09:48 |
| keesj | arm builds work just fine | 09:49 |
| keesj | I did not check on ubuntu mobile lately | 09:49 |
| Keybuk | BAD: wrong content in file 0x2a0401e0 (output), expected 'foo [#1000] (start) waiting | 09:50 |
| Keybuk | ' got 'test: foo [#1000] main process killed by SEGV signal | 09:50 |
| Keybuk | ' | 09:50 |
| Keybuk | at tests/test_initctl.c:874 (test_start_action). | 09:50 |
| Keybuk | /bin/sh: line 4: 9780 Aborted ${dir}$tst | 09:50 |
| Keybuk | FAIL: test_initctl | 09:50 |
| Keybuk | we're still waiting for porting boxes, so haven't had a chance to look at it yet | 09:51 |
| Keybuk | \o/ for test suites | 09:51 |
| Keybuk | of course, it should be noted that you communicate with Upstart 0.5.0 onwards through D-Bus | 09:54 |
| Keybuk | so you don't need the library to link to talk to it | 09:55 |
| Keybuk | the D-Bus library licence is more liberal than the GPL | 09:55 |
| keesj | http://www.paste-it.net/public/u31bbbc/ | 09:57 |
| keesj | this is what the output of initctl list looks like under qemu-arm | 09:58 |
| keesj | (0.5) that is , I was under the impression that this was just a bug | 09:59 |
| keesj | but apparently you are saying that initctl list (or status) looks nicer under x86? | 09:59 |
| keesj | or is tat 0.3.9 | 10:00 |
| Keybuk | the _5f bit? | 10:00 |
| Keybuk | that's a bug | 10:00 |
| keesj | a known one? | 10:01 |
| Keybuk | yeah | 10:02 |
| Keybuk | casey just outputs bits of dbus object paths there | 10:02 |
| Keybuk | and doesn't convert them back into names | 10:02 |
| Keybuk | (which, in all fairness, is because I didn't add the properties to the job objects) | 10:02 |
| Keybuk | right bed | 10:06 |
| keesj | :P | 10:06 |
| fbond | Hi. When entering runlevel 2, an event is emitted. Is an event emitted after the init scripts for that runlevel have finished executing? | 20:37 |
| fbond | I guess I can use "start on started rc2"? | 20:49 |
| notting | and then stopped | 20:58 |
| fbond | notting: Right, "stop on stopped rc2". | 21:02 |
| fbond | Or maybe "stop on stopping rc2" would be better... | 21:02 |
| fbond | Is there an upstart manual? | 21:02 |
| === Keybuk_ is now known as Keybuk | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!