| afournier1 | hi | 09:29 |
|---|---|---|
| afournier1 | i am trying to compile upstart 1.6.1 but configure fails on json : "Requested 'json >= 0.10' but version of json is 0.9" the problem is i cannot find json 0.10 on json-c website http://oss.metaparadigm.com/json-c/ | 09:30 |
| afournier1 | what did i miss ? | 09:30 |
| afournier1 | it got tagged 0.10 on github and they did not release the tarball :/ | 09:31 |
| afournier1 | (it's not a tag it's a branch) | 09:32 |
| xnox | afournier1: use $ pull-lp-source json-c | 09:45 |
| xnox | afournier1: or you can download a tarball from http://archive.ubuntu.com/ubuntu/pool/main/j/json-c/ | 09:45 |
| afournier1 | the thing is that i am not on ubuntu | 09:45 |
| xnox | http://archive.ubuntu.com/ubuntu/pool/main/j/json-c/json-c_0.10.orig.tar.gz | 09:46 |
| afournier1 | yes i found that one | 09:46 |
| xnox | jodh: where is json-c upstream? ^^^^ | 09:46 |
| afournier1 | i'll check the difference between http://oss.metaparadigm.com/json-c/json-c-0.9.tar.gz and http://archive.ubuntu.com/ubuntu/pool/main/j/json-c/json-c_0.10.orig.tar.gz but that's weird json-c maintener on ubuntu bump json-c version like this | 09:47 |
| jodh | afournier1: https://github.com/json-c/json-c/tags | 09:47 |
| afournier1 | ok | 09:48 |
| afournier1 | i did not see that one | 09:48 |
| afournier1 | thanks jodh | 09:48 |
| jodh | afournier1: they've hidden it rather too well ;-) | 09:48 |
| afournier1 | 7 months ago... and still no tar.gz :/ weird | 09:48 |
| jodh | afournier1: check their mailing list - I asked about it, but they apparently no longer provide releases in the old location. | 09:49 |
| xnox | afournier1: well github offers to "download" tags/revisions/etc: https://github.com/json-c/json-c/archive/json-c-0.10-20120530.tar.gz | 09:49 |
| xnox | is "auto-generated" tarball by github. | 09:49 |
| afournier1 | that's not openembedded-friendly :) | 09:51 |
| xnox | afournier1: why not? =) it's a tarball & available over https..... | 09:52 |
| afournier1 | because of the timestamp | 09:53 |
| afournier1 | but it's ok no big deal | 09:53 |
| afournier1 | oops | 09:54 |
| afournier1 | and it unpacks as json-c-json-c-0.10-20120530 :) | 09:55 |
| * xnox slowly backs away & hopes afournier1 will enjoy fiddling with it =) | 09:57 | |
| xnox | afournier1: json-c is needed for stateful re-exec. | 09:57 |
| afournier1 | you mean i should disable it ? | 09:57 |
| xnox | potentially we can make it possible to disable, if you don't care about re-execing init... but currently it's a hard requirement. | 09:58 |
| afournier1 | it's ok json-c 0.10 compiled already | 09:58 |
| afournier1 | and it's a good feature btw | 09:59 |
| felixge | stop on stopped server and stopped worker | 10:13 |
| felixge | ^--- is this supposed to work in upstart? | 10:13 |
| felixge | it seems like it doesn't work when my worker job is already stopped and I'm triggering stop on the server | 10:13 |
| xnox | jodh: stgraber: https://jenkins.qa.ubuntu.com/view/Raring/view/AutoPkgTest/job/raring-adt-upstart/5/ARCH=amd64,label=adt/artifact/results/log/*view*/ | 10:17 |
| afournier1 | yay upstart is compiled | 10:26 |
| afournier1 | btw i add to do this to json-c http://pastebin.com/1EfrZGR0 | 10:26 |
| afournier1 | i had* | 10:26 |
| jodh | stgraber: sorry - can't hear you - mumble breaking up for me. | 10:34 |
| xnox | * debian/patches/missing-headers.patch: Install json_object_iterator.h | 10:35 |
| xnox | header, needed by json.h. | 10:35 |
| xnox | afournier1: we do as well =) | 10:35 |
| stgraber | jodh: line 8691 is where it's failling when running only test_handler | 10:35 |
| afournier1 | xnox: nice to hear :) | 10:35 |
| afournier1 | xnox: now upstart does not complain beacause i don't use an initramfs, that's great | 10:36 |
| jodh | stgraber: thanks. now we play "hunt the temporary file" :-) | 10:37 |
| jodh | http://paste.ubuntu.com/1447297/ | 11:11 |
| xnox | jodh: /j #ubuntu-quality just in case ;-) | 12:20 |
| xnox | poked them there now, but no response yet. | 12:20 |
| xnox | jodh: i got pm from jibel, he will get me access to a VM later today. | 12:53 |
| * xnox already has vpn to the machines. | 12:53 | |
| felixge | I have 2 jobs that should be stopped after a third job is stopped. Is that doable with upstart? | 12:59 |
| felixge | sry, phrased this wrong | 12:59 |
| felixge | I have 2 jobs, and I want a 3rd job to stop after the 2 jobs have stopped | 12:59 |
| felixge | ^-- this | 12:59 |
| felixge | for my 3rd job I'm trying: stop on stopped job1 and stopped job2 | 13:00 |
| felixge | but this causes stop job1 to hang | 13:00 |
| felixge | (when job2 is already stopped) | 13:00 |
| xnox | felixge: is that what you really want? don't you want pre-stop script in job3 to execute stop job1; stop job2. | 13:16 |
| xnox | meh, gone? | 13:16 |
| xnox | jodh: is there a way to trace throught TEST_ALLOC_FAIL such that it would tell me which memory allocation it failed & I didn't handle right =) | 13:17 |
| abassett | hey im on ubuntu 12.04. I put a script in /etc/init, but upstart isn't recongnizing it | 13:20 |
| xnox | abassett: did you run a check on it? http://upstart.ubuntu.com/cookbook/#initctl-check-config | 13:27 |
| abassett | yea i had two problems: syntax error and i was linking to it while i worked on it | 13:28 |
| abassett | thanks | 13:28 |
| jodh | xnox: no, since the other TEST_* macros aren't aware they're being called from within such a block. You'll have to add in code to print _test_alloc_count, _test_alloc_call, and test_alloc_failed. | 13:38 |
| xnox | jodh: thanks. | 13:41 |
| xnox | jodh: your code review "uncovered" just a couple problems ;-) | 13:41 |
| jodh | xnox: better now than later ;) | 13:41 |
| felixge | another attempt: When I have a condition like this: stop on stopped jobA and stopped jobB . Will this only be satisifed if upstart sees both events? Or also if just 1 event fires and the second condition already happens to be true? | 13:44 |
| xnox | felixge: http://upstart.ubuntu.com/cookbook/#restarting-jobs-with-complex-conditions | 13:49 |
| felixge | can I met emit a "stopping" event? | 14:27 |
| felixge | initctl emit stopping someJob ? | 14:27 |
| felixge | or is the syntax: initctl emit stopping JOB=someJob ? | 14:27 |
| xnox | felixge: http://upstart.ubuntu.com/cookbook/#restarting-jobs-with-complex-conditions | 14:29 |
| xnox | felixge: it's best for you to have a pre-stop and stop on either job stopping and check if the other one is stopping or is stopped. | 14:30 |
| xnox | felixge: another way to do this is by blocking other jobs from doing stuff, see blocking in the cookbook. | 14:30 |
| felixge | xnox: thx | 14:32 |
| felixge | I think I finally figured it out | 14:32 |
| xnox | good. | 14:32 |
| xnox | =) | 14:32 |
| felixge | I have a few jobs like this: stop on stopped server and stopped worker | 14:32 |
| felixge | and now for both the server and the worker I have a pre-stop check that emits the other "stopped" event if needed | 14:32 |
| felixge | pre-stop exec status worker | grep -q "stop/waiting" && initctl emit --no-wait stopped JOB=worker || true | 14:32 |
| felixge | seems to work | 14:32 |
| xnox | =))))) | 14:33 |
| xnox | looks good. | 14:33 |
| felixge | well, it's been quite the ride :) | 14:33 |
| felixge | including a brief journey through the upstart source : ) | 14:33 |
| felixge | but that link you send me would have saved me a lot of time | 14:34 |
| felixge | I read through the upstart cookbook | 14:34 |
| felixge | and it's awesome how complete it is | 14:34 |
| felixge | but this was hard to find | 14:34 |
| felixge | :) | 14:34 |
| felixge | just soo much to read | 14:34 |
| felixge | :) | 14:34 |
| jodh | xnox/stgraber: Feel free to add more: https://wiki.ubuntu.com/FoundationsTeam/Specs/RaringUpstartUserSessions#Default_System-Provided_Session_Init_Jobs | 14:36 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!