| 00:23.27 | *** join/#asterisk-dev infobot (ibot@rikers.org) |
| 00:23.27 | *** topic/#asterisk-dev is 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 |
| 04:40.49 | *** join/#asterisk-dev CELYA_ (~Thunderbi@80.12.43.143) |
| 04:59.41 | *** join/#asterisk-dev CELYA_ (~Thunderbi@80.12.43.143) |
| 05:23.29 | *** join/#asterisk-dev CELYA_ (~Thunderbi@80.12.43.143) |
| 07:23.01 | *** join/#asterisk-dev jkroon (~jkroon@uls-154-73-35-201.wall.uls.co.za) |
| 10:37.54 | *** join/#asterisk-dev tuxian (~tuxian@194.12.3.67) |
| 12:33.31 | *** join/#asterisk-dev tsearle (~tsearle@pa-outside.zv.fraunhofer.de) |
| 12:33.31 | *** mode/#asterisk-dev [+o tsearle] by ChanServ |
| 13:25.55 | cresl1n | pokes file |
| 13:26.56 | file | Haylo |
| 13:29.24 | tsearle | poke poke |
| 13:32.18 | cresl1n | I see you tsearle |
| 14:08.03 | cresl1n | Do PBXs dream of electric CDRs? |
| 14:10.11 | file | They dream of tacos |
| 14:13.49 | seanbright | cresl1n: your presentation available online anywhere? |
| 14:15.32 | cresl1n | I think that the KamailioWorld guys are going to release videos of it |
| 14:16.07 | seanbright | cool |
| 14:20.58 | file | Needs more cowbell |
| 14:26.38 | *** join/#asterisk-dev kharwell (kharwell@nat/digium/x-vatfngdzwctyelvh) |
| 14:26.38 | *** mode/#asterisk-dev [+o kharwell] by ChanServ |
| 14:29.13 | tsearle | digital tacos & cowbells |
| 14:35.11 | *** join/#asterisk-dev bford (uid283514@gateway/web/irccloud.com/x-ztkefcaembzfmokf) |
| 14:35.11 | *** mode/#asterisk-dev [+o bford] by ChanServ |
| 14:46.54 | cresl1n | Tacos sound interesting |
| 14:47.29 | file | Always |
| 14:47.38 | file | I couldn't stand the chair anymore. |
| 14:48.46 | tsearle | @Joseph https://github.com/columbia-irt/rtptools/pull/110 |
| 14:49.04 | tsearle | @gtjoseph |
| 14:49.16 | gtjoseph | cool! |
| 14:49.37 | tsearle | did you get the test working by the way? |
| 14:50.02 | gtjoseph | yeah. i just need to add rtpdump to the test slaves |
| 14:52.20 | tsearle | did you check out the test I attached in last weeks jira too? |
| 14:52.39 | gtjoseph | no. didn't get the chance. got pulled into something else. |
| 14:52.54 | tsearle | oki |
| 15:06.03 | *** join/#asterisk-dev tuxian_ (~tuxian@igilmour.plus.com) |
| 15:27.50 | *** join/#asterisk-dev dadrc (~quassel@unaffiliated/dadrc) |
| 15:36.05 | *** join/#asterisk-dev gtjoseph (~gtjoseph@unaffiliated/gtj) |
| 15:36.05 | *** mode/#asterisk-dev [+o gtjoseph] by ChanServ |
| 15:59.14 | *** join/#asterisk-dev alex2grad (~alex2grad@38.117.100.40) |
| 16:00.24 | *** join/#asterisk-dev alex2grad (~alex2grad@38.117.100.40) |
| 16:04.19 | *** join/#asterisk-dev Worldexe (~Worldexe@95-107-33-134.dsl.orel.ru) |
| 16:09.04 | alex2grad | Hello |
| 16:09.36 | file | Hi |
| 16:10.56 | alex2grad | question about asterisk-13.21.0 and realtime pjsip: endpoins/aor/contact |
| 16:12.04 | alex2grad | if aor had qualify_frequency=0 and then qualify_frequency wad changed to 30 in realtime, the qualify wasn't run untill 'module reload res_pjsip' |
| 16:13.09 | alex2grad | I thought that aor sorcery_observer on updated should be called, but it's never called |
| 16:13.58 | gtjoseph | if it was changed in the database directly then there'd be nothing to trigger the observer |
| 16:14.25 | gtjoseph | the database isn't polled for changes |
| 16:15.25 | alex2grad | on 'pjsip show aor' it's shown a new value |
| 16:16.10 | gtjoseph | because that triggers a database fetch |
| 16:16.30 | gtjoseph | but that diesn't trigger the observers |
| 16:17.00 | alex2grad | I think how to make the observer to trigger 'updated' when sorcery object was changed in realtime |
| 16:18.37 | alex2grad | for example implement check sum in sorcery object structure and if a database fetch an object with different cehck sum that trigger observer update |
| 16:18.39 | gtjoseph | we've had some ideas on that in the past but it's not easy. we could add "on updated" triggers in the database to make a call to asterisk for instance. |
| 16:19.17 | gtjoseph | checksum might be possible as well |
| 16:19.32 | gtjoseph | th eother thing you can try is updating the contact using ARI. |
| 16:19.45 | gtjoseph | that will change the database AND trigger the observer. |
| 16:22.39 | alex2grad | could you please direct me how to implement checksum and trigger observer on update? |
| 16:24.08 | gtjoseph | well, how would you know to check the checksums? |
| 16:24.46 | gtjoseph | you don't want to do that every time a contact is looked up. |
| 16:25.02 | gtjoseph | because of the performance implications. |
| 16:27.21 | alex2grad | it's about aor, not contact |
| 16:27.50 | gtjoseph | yes aor sorry. same issues though. |
| 16:27.52 | alex2grad | but i think to modify/check checksum on every database retrives |
| 16:29.01 | alex2grad | i think to add a new global option which will manage this behavior |
| 16:30.20 | file | Having just rewritten OPTIONS support this needs to be fully defined and scoped out because it is extremely easy to tank performance |
| 16:31.03 | gtjoseph | yeah, exactly |
| 16:54.09 | *** join/#asterisk-dev jkroon (~jkroon@165.16.204.171) |
| 17:08.04 | alex2grad | is ast_sorcery_objectset_apply right place to apply checksum? |
| 17:20.20 | *** join/#asterisk-dev tuxd00d (~tuxd00d@wsip-70-166-203-100.ph.ph.cox.net) |
| 17:25.29 | *** join/#asterisk-dev malcolmd (malcolmd@pdpc/sponsor/digium/malcolmd) |
| 17:25.29 | *** mode/#asterisk-dev [+o malcolmd] by ChanServ |
| 17:37.27 | alex2grad | about checksum performance... using https://cyan4973.github.io/xxHash/ the performance degradation will be near zero |
| 17:49.33 | *** join/#asterisk-dev pchero (~pchero@176-23-78-252-cable.dk.customer.tdc.net) |
| 18:00.52 | *** join/#asterisk-dev tuxian (~tuxian@igilmour.plus.com) |
| 18:16.48 | *** join/#asterisk-dev CELYA_ (~Thunderbi@LFbn-1-11898-144.w90-93.abo.wanadoo.fr) |
| 18:21.47 | *** join/#asterisk-dev tuxd00d (~tuxd00d@wsip-70-166-203-100.ph.ph.cox.net) |
| 18:33.23 | file | the checksum method may be near zero, but the over all impact and the way it is done can have an impact |
| 18:33.30 | file | and this is not something that should be rushed into |
| 18:33.47 | file | specifically what needs to be done, how things would interact, and how it would be used should be discussed first |
| 18:50.48 | *** join/#asterisk-dev snuff-work (~snuff-wor@210.9.148.102) |
| 18:50.49 | *** mode/#asterisk-dev [+o snuff-work] by ChanServ |
| 18:57.14 | coreyfarrell | anyone have an idea about the jenkins tests/channels/pjsip/statsd/contacts failure on master? |
| 18:57.59 | coreyfarrell | thanks file |
| 18:58.24 | file | it's because of the options work, seems to be faster under master so more events go out |
| 19:05.31 | alex2grad | the goal: if realtime pjsip object is modifyed the observer should trigger 'updated' |
| 19:14.32 | alex2grad | to enable trigger we can add an extra option in sorcery.conf, for example aor=realtime,ps_aors,trigger_on_update=yes |
| 19:17.06 | alex2grad | if the trigger_on_update==yes then sorcery calculates/saves/comparies hash, if hash different then it triggers observer update |
| 19:19.02 | file | saves... where? |
| 19:22.01 | alex2grad | inside sorcery object |
| 19:22.20 | file | okay, but if the sorcery object isn't in memory - then that information is gone |
| 19:23.02 | *** join/#asterisk-dev tuxian (~tuxian@igilmour.plus.com) |
| 19:27.54 | alex2grad | I thouht the sorcery is persistence mechanism for objects. In which case the object may not be in memory? |
| 19:29.47 | file | sorcery itself doesn't guarantee that things will be in memory or are in memory |
| 19:29.52 | file | that's up to the backend implemention |
| 19:30.13 | file | there's caching ones which can be used to keep things in memory for performance reasons, but those are a configurable thing |
| 20:35.12 | alex2grad | any thoughts how to trigger 'updated' on modified realtime object? |
| 20:37.11 | alex2grad | or how to trigger pjsip_options to start/stop qualifying on updating qualify_frequency? |
| 23:33.43 | *** part/#asterisk-dev kharwell (kharwell@nat/digium/x-vatfngdzwctyelvh) |
| 23:44.22 | *** join/#asterisk-dev pchero (~pchero@176-23-78-252-cable.dk.customer.tdc.net) |