| 00:26.05 | *** part/#asterisk-dev kharwell (kharwell@nat/digium/x-adhfwctqzjicnesz) |
| 01:02.48 | rmudgett | gtjoseph: ping |
| 01:02.57 | gtjoseph | pong |
| 01:03.41 | rmudgett | concernig the clang review for sig_analog |
| 01:03.41 | gtjoseph | rmudgett: yes? |
| 01:03.45 | gtjoseph | yep |
| 01:04.32 | rmudgett | I think it would be better to just change the __analog_handle_event() idx to an int instead of changing the other function's idx to the enum |
| 01:05.22 | gtjoseph | well, i could but i'd think at least having the -1 documented as an error would be better |
| 01:05.46 | gtjoseph | nothing was preventing the array underrun before |
| 01:06.18 | gtjoseph | did you look at the latest? |
| 01:07.43 | rmudgett | It was always checking for a negative either right after calling analog_get_index or soon after in analog_hangup() |
| 01:08.21 | gtjoseph | exactly, so wouldn't it be better to have them all consistent and documented? |
| 01:08.21 | rmudgett | Yes. I looked at the latest. It just makes the diff larger to change everywhere else to the enum |
| 01:08.46 | rmudgett | They were consistently checking for negative. |
| 01:09.04 | gtjoseph | well, ok. i'll change them back |
| 01:10.22 | rmudgett | I've always been nauseated by sig_analog. Not much I can do about it since I'm not sure about some of the functionality for the various modes. |
| 01:11.08 | gtjoseph | yeah, there are some big-ass functions in there. :) |
| 01:11.35 | rmudgett | Not only that but some very old ways of doing things |
| 01:12.36 | gtjoseph | chan_unistim has a clang warning as well, but i'm not touching THAT. |
| 01:12.38 | rmudgett | What's worse is that when sig_analog was extracted, the analog code was left in chan_dahdi |
| 01:13.36 | rmudgett | Is the clang warning in chan_unistim about a zero element member? |
| 01:14.13 | gtjoseph | yes, not at the end of the structure |
| 01:14.39 | rmudgett | There's already a issue about that in JIRA. |
| 01:14.58 | gtjoseph | chan_unistim.c:1003:18: error: field 'cm' with variable sized type 'struct cmsghdr' not at the end of a struct or class is a GNU extension [-Werror,-Wgnu-variable-sized-type-not-at-end] |
| 01:15.08 | gtjoseph | same? |
| 01:15.58 | rmudgett | https://issues.asterisk.org/jira/browse/ASTERISK-25592 |
| 01:16.16 | rmudgett | yes |
| 01:16.42 | gtjoseph | i think that's totally dependent on the datagram received. i have no way to test moving it. |
| 01:17.28 | rmudgett | According to the issue, the maintainer is working on fixing it. |
| 01:18.52 | gtjoseph | I think a "#pragma GCC ingored" might do the trick :) |
| 01:35.10 | rmudgett | gtjoseph: Review looks ok to me. Is clang happy now? |
| 01:35.19 | gtjoseph | yep |
| 01:35.33 | rmudgett | +1'ed |
| 01:35.46 | gtjoseph | thanks |
| 01:36.17 | *** join/#asterisk-dev TazzNZ (~pieter@office-nat2.ytechops.co.nz) |
| 01:44.14 | gtjoseph | coreyfarrell: should testsuite pjsip pass with REF_DEBUG or are there still issues? |
| 01:47.38 | coreyfarrell | probably still many issues. my goal with pjsip was to make testable with REF_DEBUG.. so you should be able to start / stop asterisk with pjsip modules loading and not leak. can't speak much beyond that. |
| 01:48.01 | gtjoseph | gotcha |
| 01:48.12 | coreyfarrell | it's been a long time since we've had an automated REF_DEBUG run so it could be better or worse at this point |
| 01:48.27 | gtjoseph | i'm thinking worse. :) |
| 03:27.26 | *** join/#asterisk-dev leedm777 (~leedm777@50-81-28-152.client.mchsi.com) |
| 03:42.03 | *** join/#asterisk-dev leedm777 (leedm777@nat/digium/x-tcllwcynurjqswnu) |
| 04:02.09 | *** join/#asterisk-dev roramirez (~rodrigo@190.209.106.228) |
| 04:38.38 | *** part/#asterisk-dev roramirez (~rodrigo@190.209.106.228) |
| 05:44.39 | *** join/#asterisk-dev bof22 (~Thunderbi@185.13.183.107) |
| 06:25.10 | *** join/#asterisk-dev drmessano (drmessano@pdpc/supporter/active/drmessano) |
| 07:25.11 | *** join/#asterisk-dev Qwell (~north@asterisk/developer/Qwell) |
| 07:25.11 | *** mode/#asterisk-dev [+o Qwell] by ChanServ |
| 07:45.04 | *** join/#asterisk-dev Demon_VoIP (~demon@eoks.ivanovo.ru) |
| 08:03.52 | *** join/#asterisk-dev pchero_work (~pchero@109.70.54.56) |
| 08:21.56 | *** join/#asterisk-dev hehol (~hehol@gatekeeper.loca.net) |
| 08:28.51 | *** join/#asterisk-dev bof22 (~Thunderbi@185.13.183.107) |
| 08:32.07 | *** join/#asterisk-dev pchero_work (~pchero@109.70.54.56) |
| 11:34.33 | *** join/#asterisk-dev pchero_work (~pchero@109.70.54.56) |
| 11:53.52 | file | coreyfarrell, looks like it worked |
| 12:38.12 | *** join/#asterisk-dev file (~file@asterisk/developer-and-muffin-lover/file) |
| 12:38.12 | *** mode/#asterisk-dev [+o file] by ChanServ |
| 12:38.41 | *** join/#asterisk-dev moy (sid47040@gateway/web/irccloud.com/x-ksbowfmdvookfhif) |
| 13:07.20 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@216-191-106-163.dedicated.allstream.net) |
| 13:14.18 | *** join/#asterisk-dev acidfoo (~textual@modemcable002.114-70-69.static.videotron.ca) |
| 13:53.44 | *** join/#asterisk-dev leedm777 (leedm777@nat/digium/x-tqclleiptdfkoiij) |
| 14:26.23 | file | coreyfarrell, I have also scheduled a weekly run for the ref debug job now |
| 14:36.32 | *** join/#asterisk-dev goddva (~goddva@2a02:fe0:c510:e300:ae9e:17ff:fe91:a8c9) |
| 14:42.35 | *** join/#asterisk-dev cresl1n (Adium@asterisk/libpri-and-libss7-expert/Cresl1n) |
| 14:42.35 | *** mode/#asterisk-dev [+o cresl1n] by ChanServ |
| 14:43.40 | file | cresl1n, you again? |
| 14:43.47 | cresl1n | shrugs |
| 14:43.56 | cresl1n | Good morning Mr. File. |
| 14:44.02 | file | g'morning |
| 14:50.28 | cresl1n | jumps |
| 14:54.59 | *** join/#asterisk-dev putnopvut (putnopvut@asterisk/master-of-queues/mmichelson) |
| 14:54.59 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:59.09 | *** join/#asterisk-dev kharwell (kharwell@nat/digium/x-aovezihwajvxeumh) |
| 15:01.41 | *** join/#asterisk-dev mjordan (mjordan@nat/digium/x-jsfkefabgtdeyazh) |
| 15:01.41 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 15:38.02 | *** join/#asterisk-dev ehsjoar (~jonas@c-24-9-81-134.hsd1.co.comcast.net) |
| 15:39.03 | *** part/#asterisk-dev ehsjoar (~jonas@c-24-9-81-134.hsd1.co.comcast.net) |
| 15:57.33 | *** join/#asterisk-dev rmudgett (rmudgett@nat/digium/x-qcgjctnxkxnqngwo) |
| 16:53.13 | stefan27 | I'm having the weirdest bug ever. When doing a Playback(digits/16) from my asterisk, then the caller crashes, but only if the client is a linphone and only if the file played is digits/16.wav (colleagues may have tampered with the .wav files I don't know). I tested many sequences of Playback(digits/$i.wav) in randomized order and the linphone client always crashes at first Playback(digits/16). |
| 16:53.13 | stefan27 | How can I check if a soundfile has been corrupted somehow? I dont understand why other clients hear "16" just fine. |
| 16:56.31 | stefan27 | (sorry this should have gone to #asterisk in retrospect) |
| 16:58.27 | rmudgett | That really looks like a problem with linphone itself |
| 17:05.18 | *** join/#asterisk-dev jameswf (~jameswf@unaffiliated/jameswf-home) |
| 17:07.21 | jsmith | stefan27: While that's probably a bug in linphone itself, you might try with the new (1.5) release of asterisk-sounds and see if that improves things |
| 17:08.15 | stefan27 | I guess you're right. I tried opening the 16.wav files and removing some random bytes at the end and everything works |
| 17:08.58 | stefan27 | but superweird, i've done hundreds of playbacks of wav files on various linphones and never any issue |
| 17:12.47 | stefan27 | Maybe linphone has an extremely obscure bug which requires very specific timings or circumstances. thanks for your answers it's not an asterisk-dev-concern... |
| 18:37.40 | coreyfarrell | gtjoseph: results from the REF_DEBUG look close to what I remember, about 180 of the PJSIP tests leaked refs, plus about 30 rest_api tests. some ARI tests may use PJSIP but ARI has leaks on it's own. |
| 18:38.10 | gtjoseph | Ok, that matches what I saw. Thanks for checking. |
| 18:38.47 | coreyfarrell | thank file, he's the one that got the jenkins job working :) |
| 18:39.18 | gtjoseph | thanks file! |
| 18:39.40 | file | if there's any other jobs that need doing or tweaks, if you make them I can deploy them now |
| 18:39.51 | file | knows the stuff and things |
| 18:40.14 | file | there's also realtime tests now, woot |
| 18:40.25 | gtjoseph | so what kind of alembic checking is happening now? |
| 18:41.00 | file | the check pipeline makes sure there's no branches, and nightly realtime tests are done against PJSIP which include running through alembic completely |
| 18:41.11 | coreyfarrell | file: the only tweak I'd like to see at this point is to generate gcov reports |
| 18:41.11 | gtjoseph | cool! |
| 18:41.43 | file | putnopvut wrote a script which converts the pjsip.conf into realtime, so any future PJSIP tests will also be run against realtime |
| 18:43.00 | file | coreyfarrell, ah |
| 18:43.15 | file | next on my personal list is to get the asterisk gates working, but that's a weekend thing |
| 18:48.03 | *** join/#asterisk-dev Demon_VoIP (~demon@ip253.net222.n37.ru) |
| 19:07.17 | *** join/#asterisk-dev seanbrig1t (~sean@asterisk/contributor-and-bug-marshal/seanbright) |
| 19:07.17 | *** mode/#asterisk-dev [+o seanbrig1t] by ChanServ |
| 19:07.58 | *** join/#asterisk-dev seanbrig1t (~sean@asterisk/contributor-and-bug-marshal/seanbright) |
| 19:07.58 | *** mode/#asterisk-dev [+o seanbrig1t] by ChanServ |
| 19:14.44 | *** join/#asterisk-dev wdoekes (~walter@wjd.osso.nl) |
| 19:21.43 | *** join/#asterisk-dev seanbright (~sean@asterisk/contributor-and-bug-marshal/seanbright) |
| 19:21.43 | *** mode/#asterisk-dev [+o seanbright] by ChanServ |
| 19:28.36 | *** join/#asterisk-dev luckman212 (~luckman21@unaffiliated/luckman212) |
| 19:48.59 | *** join/#asterisk-dev goddva (~goddva@2a02:fe0:c510:e300:ae9e:17ff:fe91:a8c9) |
| 21:07.46 | *** join/#asterisk-dev luckman212 (~luckman21@unaffiliated/luckman212) |
| 21:18.46 | *** join/#asterisk-dev mjordan (mjordan@nat/digium/x-siddeinrgmxjtshs) |
| 21:18.46 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 21:24.19 | gtjoseph | hey file... was testing sorcery memory cache realtime... is it normal for the objects to not show up in the cli when listed? |
| 21:24.28 | gtjoseph | cache with realtime |
| 21:25.08 | file | listed using what? |
| 21:25.18 | file | and which caching model? |
| 21:25.20 | gtjoseph | pjsip list ... |
| 21:25.41 | gtjoseph | copied from wiki "memory_cache,maximum_objects=150,expire_on_reload=yes,object_lifetime_maximum=3600" |
| 21:26.03 | gtjoseph | they show if retrieved explicitly |
| 21:26.14 | gtjoseph | pjsip show transport abc |
| 21:26.22 | file | the listing should pull from the realtime |
| 21:26.26 | file | individual request pulls from cache |
| 21:26.36 | gtjoseph | hmmm |
| 21:26.44 | gtjoseph | does order in sorcery.conf matter |
| 21:26.53 | file | cache has to be first |
| 21:27.14 | gtjoseph | it's before realtime but after config |
| 21:27.36 | file | I haven't done it with both config and realtime |
| 21:27.50 | gtjoseph | oh interesting |
| 21:27.57 | gtjoseph | it has to be before all others |
| 21:28.06 | gtjoseph | cache, config, realtime works |
| 21:28.22 | gtjoseph | might be worth a note |
| 21:34.42 | gtjoseph | why no retrieve multiple support? Too much work to test each object? |
| 21:38.03 | file | because unless you enable full backend caching, then it's a subset |
| 21:38.23 | file | so you have to query the backend regardless |
| 21:39.05 | gtjoseph | how would it be not full? |
| 21:39.27 | file | in non-full it caches upon use |
| 21:39.36 | file | if you don't use some, then they aren't cached |
| 21:39.48 | gtjoseph | yeah, ok. |
| 21:40.29 | gtjoseph | there are places that do retrieve_by_fields on transports so they can find one that matches... mat, endpoint_identified, etc |
| 21:40.34 | gtjoseph | nat |
| 21:40.42 | gtjoseph | identifier |
| 21:41.00 | gtjoseph | and i was trying to prevent having to go through the full apply process every time |
| 21:46.52 | gtjoseph | file: full_backend_cache isn't documented on the wiki. i just now realized what you were referencing |
| 21:47.20 | *** join/#asterisk-dev luckman212 (~luckman21@unaffiliated/luckman212) |
| 21:47.26 | file | hrm |
| 21:48.34 | file | ah |
| 21:48.38 | file | makes note |
| 21:50.12 | gtjoseph | and with full_backend_cache, realtime now works great, but i can't list from config. :) |
| 21:52.15 | coreyfarrell | the main screen of jenkins has "Last Success" / "Last Failure" columns. success says the job ran to completion, even if some tests failed? |
| 21:57.19 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 22:08.56 | *** join/#asterisk-dev leedm777 (~leedm777@50-81-28-152.client.mchsi.com) |
| 22:14.00 | *** join/#asterisk-dev mjordan (mjordan@nat/digium/x-aiubhickydnnbzee) |
| 22:14.00 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 22:16.18 | *** join/#asterisk-dev [TK]D-Fender (~joe@64.235.216.2) |
| 22:28.20 | *** join/#asterisk-dev cresl1n (Adium@asterisk/libpri-and-libss7-expert/Cresl1n) |
| 22:28.20 | *** mode/#asterisk-dev [+o cresl1n] by ChanServ |
| 22:43.58 | *** join/#asterisk-dev leedm777 (leedm777@nat/digium/x-przicigcelehwzyl) |
| 22:44.20 | *** join/#asterisk-dev cresl1n (~Adium@asterisk/libpri-and-libss7-expert/Cresl1n) |
| 22:44.20 | *** mode/#asterisk-dev [+o cresl1n] by ChanServ |
| 23:19.59 | *** join/#asterisk-dev roramirez (~rodrigo@190.209.106.228) |
| 23:23.34 | roramirez | I see the dbdel function is deprecated around 2006. What is the process for remove deprecated function. Want to continue retain dbdel function? |
| 23:42.06 | file | it's not doing any harm being there, so I'd keep it |