| 01:08.33 | *** join/#asterisk-dev seanbright_ (~sean@asterisk/community-developer/seanbright) |
| 01:08.33 | *** mode/#asterisk-dev [+o seanbright_] by ChanServ |
| 02:06.54 | *** join/#asterisk-dev jkroon (~jkroon@165.16.204.101) |
| 02:06.54 | *** mode/#asterisk-dev [+o jkroon] by ChanServ |
| 02:34.38 | *** join/#asterisk-dev jkroon (~jkroon@165.16.204.101) |
| 02:34.38 | *** mode/#asterisk-dev [+o jkroon] by ChanServ |
| 05:57.22 | *** join/#asterisk-dev infobot (ibot@96-86-209-99-static.hfc.comcastbusiness.net) |
| 05:57.22 | *** topic/#asterisk-dev is Take the March 2019 Asterisk User Survey! https://goo.gl/forms/xL1VUHRsf95saly13 -- Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- Check out our blog! blogs.asterisk.org -=- Follow on Twitter at @AsteriskDev |
| 07:08.20 | *** join/#asterisk-dev jkroon (~jkroon@165.16.204.110) |
| 07:08.20 | *** mode/#asterisk-dev [+o jkroon] by ChanServ |
| 07:12.56 | *** join/#asterisk-dev jkroon (~jkroon@165.16.204.110) |
| 07:12.56 | *** mode/#asterisk-dev [+o jkroon] by ChanServ |
| 08:28.29 | *** join/#asterisk-dev jkroon (~jkroon@165.16.204.110) |
| 08:28.30 | *** mode/#asterisk-dev [+o jkroon] by ChanServ |
| 10:47.01 | *** join/#asterisk-dev jkroon_ (~jkroon@165.16.204.109) |
| 10:47.01 | *** mode/#asterisk-dev [+o jkroon_] by ChanServ |
| 11:55.54 | *** join/#asterisk-dev AsteriskRoss (~AsteriskR@37.157.48.2) |
| 13:33.12 | jkroon | so after two days of struggling, we're again downgrading our core nodes from 13 to 11. |
| 13:33.30 | jkroon | without pjsip, are there any known RTP corruption issues? |
| 13:35.32 | seanbright_ | jira would be the source of truth for that question i'd assume |
| 13:38.43 | jkroon | not finding anything there. |
| 13:38.54 | jkroon | well, we'll figure out a different way of testing. |
| 13:39.07 | jkroon | but for now I can't focus with all the people cussing at us. |
| 13:41.59 | seanbright_ | what is 'rtp corruption' exactly? |
| 13:42.06 | seanbright_ | poor sound quality? |
| 13:42.54 | jkroon | rtp frames comes in one side, when decoded (wireshark) from these frames, they sound correct, decoding the egress rtp frames and playing back sounds like garbled thunder or something. |
| 13:43.17 | jkroon | seems to mostly happen if one of the ends switches SSRC or have marks in the RTP stream. |
| 13:44.30 | seanbright_ | well there have been various changes around those two things in the last few weeks |
| 13:44.39 | seanbright_ | not in 13 though |
| 13:44.39 | jkroon | We'll figure it out, there are a few IAX/2 channels involved in some of the cases too, so we're wondering if that's not contributing. In some cases we're seeing massive rtp timestamp jumps without the mark bit ... i was initially thinking remote side is choking on thta. |
| 13:44.58 | seanbright_ | oh |
| 13:45.04 | seanbright_ | you lost me at IAX2 |
| 13:45.07 | seanbright_ | :D |
| 13:45.51 | seanbright_ | i've also seen bugs that mention timestamp skew |
| 13:45.57 | jkroon | seanbright_, that's good to know ... yea, so one option for us (which I'm pro) is to re-run the UAT against asterisk 16, but first we need an exact reproducible case for this, which we think we've made progress on. and last two weeks is not yet in release. |
| 13:46.49 | jkroon | yea, i patched asterisk (not in this specific case) in some cases to "automagically" set the mark bit if it detects timestamp jumps > 1000ms ... which solved a few similar problems for us. |
| 13:47.31 | seanbright | i think we were even talking about this in here a few days ago |
| 13:48.04 | jkroon | i missed that convo. i think I raised the lack of change of SSRC at some point, but suspect that relates mostly to PJSIP. |
| 13:48.17 | jkroon | also not setting the mark bit. |
| 13:48.48 | seanbright | hmm |
| 13:48.50 | jkroon | and I don't know RTP well enough to know exactly when what SHOULD/MUST happen, so this is quite hard for us (seeing that I'm lead here) |
| 13:49.20 | jkroon | sorry, change of SSRC when the actual source of RTP changes, eg, from a local playback to a remote incoming source. |
| 13:49.29 | jkroon | but I'm barfing from memory here ... |
| 13:49.48 | jkroon | and IAX/2 I believe doesn't have the concept of a SSRC, but it does have the mark bit. |
| 13:51.31 | jkroon | let me first roll back, get a good nights sleep and start building a test env in the morning. Will need to get our peering providers to provide some assistance. |
| 13:51.34 | seanbright | well, i know there have been changes that have gone in recently in regards to SSRC changes, but i am not able to look for it right now |
| 13:52.40 | jkroon | seanbright, thanks, just knowing that already helps thanks, it gives me something to start searching for in git, and just that knowledge is already further than I was prior to this discussion, so thank you. |
| 14:49.00 | *** join/#asterisk-dev bford (uid283514@gateway/web/irccloud.com/x-awcigqskonjhmgtk) |
| 14:49.00 | *** mode/#asterisk-dev [+o bford] by ChanServ |
| 14:51.05 | *** join/#asterisk-dev AsteriskRoss (~AsteriskR@37.157.48.2) |
| 16:43.45 | *** join/#asterisk-dev offbeatwitch (~offbeatwi@2001:bc8:608:131::1) |
| 17:06.43 | jkroon | on https://gerrit.asterisk.org/c/asterisk/+/15628, gtjoseph referenced a bug, i added to the commit message, but i'm not seeing it creating the link with https://issues.asterisk.org/jira/browse/ASTERISK-29348 - what am I missing? |
| 17:07.03 | file | the topic has to be set as well |
| 17:07.08 | jkroon | ah |
| 17:07.47 | jkroon | done, thanks. |
| 17:09.27 | jkroon | file, whilst you're here, re the deprecated stuff, most people only ever look at the logs if there are problems. how many people do you reckon will see those error messages/warnings? Perhaps add something to the build as well, so at the end of make, if there are deprecated modules even just compiled, warning at that point: you've built the following deprecated modules: ... ... ... |
| 17:09.48 | file | I don't think I can sanely do that there |
| 17:10.20 | jkroon | <PROTECTED> |
| 17:10.39 | file | but I'll accept further changes to make it visible, as it is we're giving notice in the release notes/changelog, menuselect, and console |
| 17:10.41 | file | Makefile |
| 17:10.41 | jkroon | menuselect has the option, so perhaps a menuselect --output-deprecated ? |
| 17:10.56 | jkroon | has the *information |
| 17:12.18 | jkroon | file, just an idea, if I ever feel bored again and looking to tamper I can definitely take a look but not making promises. |
| 17:12.53 | jkroon | i'm assuming this already also reflects on cli for modules show? |
| 17:13.07 | file | not currently, it could |
| 17:13.24 | file | such things were not originally brought up last year |
| 17:13.32 | jkroon | there are lots of things that *could* ... just a bunch of effort :). |
| 17:13.55 | jkroon | anyway, get the basics, leave the rest for people that gets bitten by not having seen the messages |
| 17:14.04 | file | people will ignore it at all regardless |
| 17:14.06 | jkroon | either they'll learn to read logs, or file a patch :) |
| 17:14.08 | file | er ignore it all |
| 17:14.16 | seanbright | um, i think you mean irregardless |
| 17:14.23 | jkroon | they only need to get bitten *once* |
| 17:14.40 | file | seanbright: get off my lawn!!! |
| 17:15.11 | seanbright | heh |
| 17:15.30 | file | I wrote an email to the grocery store yesterday complaining, I felt very "get off my lawn" old manish |
| 18:02.48 | *** join/#asterisk-dev pchero (~pchero@211.178.226.108) |
| 18:02.48 | *** mode/#asterisk-dev [+o pchero] by ChanServ |
| 18:11.24 | *** join/#asterisk-dev offbeatwitch (~offbeatwi@platypus.offbeatwit.ch) |
| 22:42.06 | *** join/#asterisk-dev jkroon (~jkroon@165.16.204.109) |
| 22:42.06 | *** mode/#asterisk-dev [+o jkroon] by ChanServ |