Minor update to the wrapper_script portion to compensate for the pipe that can fail anywhere. Reproducible: Always Steps to Reproduce:
Created attachment 219799 [details, diff] init script patch This patch needs to be applied to the startup script. It enforces pipefail on the asterisk | logger pipeline so that if asterisk crashes it can once more get restarted properly instead of thinking that asterisk shut down properly (as per logger that cleans out properly).
This should have been merged a long long time ago already.
Created attachment 239385 [details, diff] asterisk.initd3.patch Re-adds logger statement for which pipefail was required. As per IRC discussion: <Chainsaw> Yeah, the structure's a little bit different. <Chainsaw> I wanted to just apply it, but it wasn't clear to me that it still made sense. <Chainsaw> I'm leaning towards saying "we no longer require this". <Chainsaw> (Because of whatever else I applied since then) <jkroon> well, i put that logger in there for a reason. <jkroon> supposedly to trap any stderr output and get it logged properly. <jkroon> so if you feel that that's not required then the ebuild "as is" is fine and "we no longer require this" is an appropriate close ... your call
+*asterisk-1.6.2.9-r1 (19 Jul 2010) + + 19 Jul 2010; <chainsaw@gentoo.org> +files/1.6.2/asterisk.initd, + +files/1.6.2/call_data.txt, +files/1.6.2/find_call_ids.sh, + +files/1.6.2/find_call_sip_trace.sh, +files/1.6.2/sip_calc_auth, + +asterisk-1.6.2.9-r1.ebuild: + Revision bump for enhancements by Jaco Kroon in bugs #300832, #305265 & + #328517. This improves the initscript and simplifies the ebuild. + Environmental variables for module overrides (which were a bad idea in the + first place) are no longer supported.
