| 01:36.46 | *** join/#asterisk-dev electronic_eel_ (~quassel@213.240.182.14) |
| 03:42.25 | *** join/#asterisk-dev electronic_eel (~quassel@213.240.182.36) |
| 11:56.14 | *** join/#asterisk-dev n0tiz (~n0tiz@82-69-15-38.dsl.in-addr.zen.co.uk) |
| 12:29.37 | *** join/#asterisk-dev n0tiz (~n0tiz@82-69-15-38.dsl.in-addr.zen.co.uk) |
| 13:51.33 | *** join/#asterisk-dev Dovid (~dovid@ool-4356e81f.dyn.optonline.net) |
| 13:51.59 | Dovid | Happy new year everyone |
| 13:52.20 | Dovid | Does Asterisk get "kranky" when it does not get packets that it expects? |
| 14:05.37 | file | that query is not specific enough - each receiver of packets is its own implementation, what it does with invalid packets is up to it but they should all tolerate them |
| 14:29.54 | Dovid | file: I signed up for the dev list since I think it's more appropiate for dev then it is for the users list. Is there a manual approval process? |
| 14:30.03 | file | I do not know |
| 14:30.50 | file | I can log in and see if anything is pending |
| 14:31.07 | Dovid | I am going to wait till I get approved so I don't spam both lists. This is what is in my email: |
| 14:31.09 | Dovid | https://paste.ubuntu.com/p/bBhV3WRpjt/ |
| 14:32.07 | file | there is no pending membership for you |
| 14:33.21 | file | your question also involves something which is not commonly used in a way that is not commonly done, so expect any responses to reflect that |
| 14:33.50 | file | and to be pedantic - it's res_musiconhold |
| 14:33.54 | Dovid | file: Strange. I will try again from a different browser. |
| 14:34.31 | Dovid | meaning it's old code that isn't touched much? |
| 14:34.52 | file | it's old code that isn't touched much, and that few people use in the way you're doing |
| 14:35.12 | file | thus few people familiar with it, and I think it even predates my involvement |
| 14:36.54 | Dovid | file: I know I had issues with local channels playing sounds and when I removed local channels it got a lot better. In this case it's MOH but I have other load and CPU issues that can't be explained (not with MOH) and I am wondering if Asterisk is not happy with not getting packets and if thereis some bug in asterisk with how it handles packets. I am using chan_sip. I doubt pjsip will change it but worth a shot. |
| 14:37.38 | file | each consumer of packets handles them in their way |
| 14:37.47 | file | the core doesn't care |
| 14:38.19 | file | the specific functionality in res_musiconhold you use may not like something |
| 14:39.13 | file | without reproducing, digging in, it's all just guesses |
| 14:39.35 | Dovid | file: OK. So what your saying is it's more likely an issue between ffmpeg and res_musiconhold then res_musiconhold and the core |
| 14:39.43 | file | yes. |
| 14:40.18 | Dovid | file: Any comment on this one? http://lists.digium.com/pipermail/asterisk-users/2020-December/295595.html |
| 14:40.41 | file | there's no comment I can give |
| 14:41.02 | file | without digging in and specifically diagnosing your question is the equivalent of going "this thing randomly turns on, why does it do that?" |
| 14:41.04 | file | of which I can not say |
| 14:41.27 | file | that's the thing with diagnosing this kind of stuff - surface level information generally yields little |
| 14:43.17 | Dovid | I am by no means a dev but is there any way to quanify which part/process of Asterisk are going on so we can tell what may be causing that spike? A debug does not show much. Know anyone that can diagnose such issues? Can we pay Sangoma to daignose it? |
| 14:43.59 | file | Asterisk doesn't keep track of CPU consumption of individual modules/components/etc, you generally have to eliminate things further and further or add explicit code to disable parts and isolate |
| 14:44.47 | file | I don't know of anyone, and I don't know if that would be something we would take on |
| 14:44.55 | file | you could try contacting sales |
| 14:45.02 | Dovid | thanks. |
| 14:45.11 | Dovid | have a great weekend and a happy new year |
| 19:49.10 | *** join/#asterisk-dev sh_smith (~sh_smith@cpe-172-88-21-24.socal.res.rr.com) |