| 00:01.07 | gtjoseph | file: still there? |
| 00:01.18 | file | barely |
| 00:02.31 | gtjoseph | :) the change to ast_sorcery_is_object_field_registered worked fine but it has to compile a regex for each compare so i was thinking about adding a regex_t to ast_sorcery_object_field and compiling the regex only once in ast_sorcery_object_fields_register |
| 00:02.45 | file | ermph |
| 00:03.32 | gtjoseph | did you just fall off the couch? |
| 00:04.16 | file | maybe |
| 00:05.31 | file | I have no alternative |
| 00:06.46 | gtjoseph | actually the only use of ast_sorcery_object_fields_register is me in phoneprov and sorcery to process the automatic ^@ fields so it should be low impact. |
| 00:07.07 | gtjoseph | not to mention saving some cycles and memory |
| 02:07.33 | *** join/#asterisk-dev mjordan (~mjordan@75.76.55.191) |
| 02:07.33 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 02:59.27 | *** join/#asterisk-dev mjordan (~mjordan@75.76.55.191) |
| 02:59.27 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 07:12.57 | *** join/#asterisk-dev infobot (ibot@rikers.org) |
| 07:12.57 | *** topic/#asterisk-dev is Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- Asterisk 13.0.0 now released. Do the release dance! -=- Follow on Twitter at @AsteriskDev |
| 07:18.56 | *** join/#asterisk-dev infobot (ibot@rikers.org) |
| 07:18.56 | *** topic/#asterisk-dev is Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- Asterisk 13.0.0 now released. Do the release dance! -=- Follow on Twitter at @AsteriskDev |
| 07:48.39 | *** join/#asterisk-dev infobot (ibot@rikers.org) |
| 07:48.39 | *** topic/#asterisk-dev is Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- Asterisk 13.0.0 now released. Do the release dance! -=- Follow on Twitter at @AsteriskDev |
| 12:49.20 | *** join/#asterisk-dev mjordan (~mjordan@75.76.55.191) |
| 12:49.20 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 14:14.29 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@64.235.216.2) |
| 14:30.04 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@64.235.216.2) |
| 14:43.18 | *** join/#asterisk-dev mjordan (~mjordan@75.76.55.191) |
| 14:43.18 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 15:05.58 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-uynnibisgmtuyaep) |
| 15:05.58 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 16:10.58 | *** join/#asterisk-dev gtjoseph (~gtj0@unaffiliated/gtj) |
| 18:49.58 | *** join/#asterisk-dev moy (uid47040@gateway/web/irccloud.com/x-xicaniffaczpvznl) |
| 18:53.19 | *** join/#asterisk-dev leifmadsen (~leifmadse@asterisk/documenteur-extraordinaire/blitzrage) |
| 18:53.19 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 18:55.01 | *** join/#asterisk-dev jrose_atDigium (~jrose_atD@nat/digium/x-rkrhedyuwpgfbypg) |
| 18:56.01 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-nfwvfwvzbqsafnff) |
| 18:56.01 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 19:01.10 | *** join/#asterisk-dev X-Rob_ (sid14615@gateway/web/irccloud.com/x-nftcxmvsrwhoxfvf) |
| 19:07.23 | *** join/#asterisk-dev elguero (~miguel323@2001:470:1f06:12c4::2) |
| 19:34.05 | coreyfarrell | mjordan: ping |
| 19:34.55 | file | coreyfarrell, I found a thing - I think you and putnopvut were talking about it |
| 19:35.02 | file | https://issues.asterisk.org/jira/browse/ASTERISK-24020 |
| 19:35.14 | skrusty | evening |
| 19:38.55 | coreyfarrell | file: that definitely looks like the issue. I just discarded the review for fixing the testsuite test that exposed the bug |
| 19:40.03 | coreyfarrell | oh wait, that bug is for the leaked bridge reference that I recently fixed in asterisk |
| 19:40.32 | file | for AMI Bridge? |
| 19:41.00 | coreyfarrell | yeah... bridge_action wasn't doesn't ao2_cleanup for the allocation reference, so the bridge would never free.. that's fixed |
| 19:41.08 | file | cool |
| 19:41.12 | coreyfarrell | wasn't doing that is |
| 19:41.14 | file | you can close it |
| 19:41.35 | file | or me, but since you did it :D |
| 19:41.37 | coreyfarrell | new issue we're having is in tests/manager/bridge_action, one of the channels doesn't get hung up |
| 19:43.02 | coreyfarrell | I'll close it |
| 19:45.13 | file | oh he mentioned PJSIP - that's why I saw it |
| 19:45.25 | coreyfarrell | so putnopvut and I were talking about another issue with bridge_action |
| 19:46.01 | skrusty | is there a way to set things like from domain when creating a new channel in ari? |
| 19:46.27 | coreyfarrell | there is some situation where channel A & B are bridged, when channel B hangs up it causes channel A to go back to the extension/priority it originally came from |
| 19:46.47 | file | skrusty, no - you have to create an explicit endpoint |
| 19:46.54 | skrusty | hmm :/ |
| 19:47.07 | coreyfarrell | one of these days I will learn/understand the bridging code, not today though :( |
| 19:49.41 | skrusty | bum |
| 19:49.52 | coreyfarrell | file: I just discarded https://reviewboard.asterisk.org/r/4166/ but I'm interested in what you think of the issue we were discussing.. I'll search JIRA see if someone else has reported |
| 19:49.58 | skrusty | i was hoping that there might be some way of doing to by setting chan vars or something :( |
| 19:52.42 | file | coreyfarrell, I'm about to head out to dinner - I'll look when I get there |
| 19:52.44 | file | skrusty, not presently |
| 19:52.52 | coreyfarrell | file: awesome thanks |
| 19:52.57 | coreyfarrell | enjoy dinner |
| 19:59.50 | *** join/#asterisk-dev tonyclewis_ (sid6025@gateway/web/irccloud.com/x-ooyyqmllurcflqxe) |
| 20:05.19 | *** join/#asterisk-dev zerohalo (~zerohalo@2601:6:f80:224:a59e:6b02:8552:f94) |
| 20:07.49 | *** join/#asterisk-dev jrose_atDigium (~jrose_atD@nat/digium/x-rkrhedyuwpgfbypg) |
| 20:07.49 | *** join/#asterisk-dev tm1000 (uid6728@gateway/web/irccloud.com/x-hoavtegijlbhujag) |
| 20:07.49 | *** join/#asterisk-dev GameGamer43 (sid5533@gateway/web/irccloud.com/x-quhpjtpjcfxousua) |
| 20:11.33 | *** join/#asterisk-dev luckman212 (~luckman21@unaffiliated/luckman212) |
| 20:15.38 | *** join/#asterisk-dev vlad_starkov (~vlad_star@91.206.59.50) |
| 20:45.23 | *** join/#asterisk-dev mycent1 (d932e7cb@gateway/web/cgi-irc/kiwiirc.com/ip.217.50.231.203) |
| 20:45.55 | mycent1 | hey guys :) i try to connect my twillio to asterix to resive calls on my voip |
| 20:47.00 | mycent1 | voip on 2 softphones each other is working great , but if i try to send inbound its allways "no matichung endpoint" , how can i set up, that all inbound is ring on all softphones ? |
| 20:47.22 | mycent1 | wrong channel |
| 20:59.41 | file | coreyfarrell, I err with Mark at first glance |
| 21:00.38 | coreyfarrell | file: I agree Mark is right, it's a problem with Asterisk.. I was asking if you had any ideas :) |
| 21:00.49 | file | nope |
| 21:01.07 | file | I have not spent much time in that specific area |
| 21:01.43 | coreyfarrell | one thing I was thinking is the first time a channel is bridged, the channel is running inside __ast_pbx_run, so ast_set_flag(ast_channel_flags(c), AST_FLAG_IN_AUTOLOOP); has been run |
| 21:02.14 | coreyfarrell | but when you do Action: Bridge to move a channel from one bridge to another, __ast_pbx_run has finished and the flag is unset |
| 21:02.38 | coreyfarrell | thus it fails to bump the return priority |
| 21:03.41 | coreyfarrell | I admit though I don't really know what happens to the channel in relation to pbx.c when it is bridged |
| 21:04.11 | file | nothing whilst it is bridged |
| 21:08.01 | coreyfarrell | so before bridging, Echo() is running with __ast_pbx_run in the backtrace.. does that thread just pause while the channel is bridged? |
| 21:09.16 | file | no, you can't do two things at once on a channel - if it's not already in bridging then the channel has to be masqueraded away |
| 21:10.42 | coreyfarrell | ok, so the thread that was running Echo() exits, which means __ast_pbx_run would return.. so I think I understand what's causing the issue |
| 21:54.43 | gtjoseph | mjordan: you really there? |
| 22:00.11 | mjordan | gtjoseph: yes-ish. Have to walk the dog, but I'll be back for a bit in about 20 min |
| 22:00.35 | gtjoseph | no worries, just had a quick testsuite question for you |
| 22:15.04 | mjordan | gtjoseph: back |
| 22:15.47 | gtjoseph | so, do you have any idea what the maximum amount of time it takes a single testsuite test to run on your test boxes? |
| 22:17.05 | gtjoseph | mjordan: no need to look or anything, just a guess if you have one |
| 22:19.03 | mjordan | yup |
| 22:19.06 | mjordan | looks |
| 22:19.58 | mjordan | 151 min is the last full run on a 'reasonable' machine |
| 22:20.08 | mjordan | 188 min on the slow machine |
| 22:20.36 | gtjoseph | that's for the whole run? Any idea what the longest individual test is? |
| 22:28.35 | mjordan | hm. |
| 22:28.48 | mjordan | well, the longest running tests are the voicemail, fax, and confbridge tests |
| 22:28.52 | mjordan | by a long shot |
| 22:29.02 | mjordan | you can run the important bits much quicker |
| 22:29.05 | gtjoseph | and fastagi/record which hangs forever :) |
| 22:29.08 | mjordan | er |
| 22:29.10 | mjordan | eek? |
| 22:29.17 | gtjoseph | for me aand a few others anyway |
| 22:29.28 | mjordan | I'll take a look at it. Those are pretty old tests. |
| 22:29.36 | coreyfarrell | yeah that happens to me |
| 22:29.42 | mjordan | basically, tests in channels should run relatively quickly |
| 22:29.45 | gtjoseph | anyway...I'm adding a --timeout parameter to runtests.py which will abort a test if it's hung. |
| 22:29.52 | mjordan | that would be nice |
| 22:29.54 | coreyfarrell | seems like it's more frequent if I haven't restarted my computer in a while |
| 22:30.01 | mjordan | how are you determining its 'hungness'? |
| 22:30.37 | coreyfarrell | asterisk process at 0% CPU usage, testsuite logs say that reactor has timed out, but it doesn't stop asterisk or the test |
| 22:30.41 | mjordan | coreyfarrell: btw, spent most of the day cleaning up the last of the rest_api failures. |
| 22:30.45 | gtjoseph | --timeout 60 would abort a test if it hasn't produced any output in 60 seconds. The default is inf so today's behavuir. |
| 22:30.53 | mjordan | that would work for me |
| 22:31.06 | mjordan | most tests should, even if they are resetting the reactor, still output something |
| 22:31.13 | gtjoseph | yep |
| 22:31.22 | mjordan | looks at something |
| 22:31.29 | coreyfarrell | resetting the reactor timeout produces a log message |
| 22:32.06 | gtjoseph | yep. only a really hung test would trip the timeout. |
| 22:32.07 | mjordan | right, but it's only at INFO |
| 22:32.24 | mjordan | and our default streamhandler in logger.conf only emits WARN+, which means runtests.py wouldn't see it |
| 22:32.37 | mjordan | you could make logger.conf emit INFO+ by default |
| 22:32.56 | coreyfarrell | mjordan: that's awesome.. I have a review that was approved to clear up a bunch of the bridges not being deleted, but then a change was committed that removed the StasisEnd watchers, so my approach is probably mostly bad |
| 22:33.28 | mjordan | coreyfarrell: I'm just going to commit it - been running it in a loop all day (yay pretty_print) and it seems to have resolved all the failure for me |
| 22:33.38 | mjordan | but the 64-bit build agent may prove me false ;-) |
| 22:34.22 | mjordan | gtjoseph: I definitely like that idea. We still have the testsuite get hung up on the build agents, albeit rarely |
| 22:34.29 | file | mjordan, commit it all! |
| 22:34.33 | coreyfarrell | yeah the change I was making was mostly for tests causing leaks, not really for tests that failed to hangup channels |
| 22:34.37 | file | anything goes on the weekend |
| 22:35.33 | mjordan | weekend commits are the best |
| 22:35.39 | mjordan | time to clean up any screw ups you make :-D |
| 22:35.42 | coreyfarrell | something I just recently realized, when a test doesn't hangup all channels and we end up doing kill -9 $asterisk_pid, coverage doesn't get any stats for that test |
| 22:36.11 | coreyfarrell | mjordan: speaking of screw ups I made, did you notice r4186? |
| 22:36.16 | mjordan | yup, no worries |
| 22:36.27 | mjordan | we'll do a regression release when it's time |
| 22:36.37 | mjordan | for 1.8 as well :-) |
| 22:36.49 | file | 1.8? what's that? |
| 22:37.00 | mjordan | file: already paged that memory out? |
| 22:37.07 | file | mjordan, yes |
| 22:37.08 | coreyfarrell | sucks I already (happily) deleted by 1.8 working copy |
| 22:37.12 | mjordan | lol |
| 22:37.22 | mjordan | on my last laptop I still had a 1.4 checkout |
| 22:37.22 | gtjoseph | mjordan: https://reviewboard.asterisk.org/r/4145/ updated. :) |
| 22:37.27 | mjordan | Because you never know! |
| 22:37.55 | mjordan | gtjoseph: cool, I'll take a look at it tonight |
| 22:38.00 | mjordan | darn evening plans |
| 22:38.04 | coreyfarrell | I think I used 1.4 for like an hour |
| 22:38.10 | mjordan | going to take me away from a nice evening of coding and testing |
| 22:38.17 | mjordan | coreyfarrell: burn! |
| 22:39.20 | coreyfarrell | also gives away how long I've been at this.. 1.6 was the only standard release I've ever used |
| 22:39.43 | mjordan | I came into the project at 1.8.5. :-P |
| 22:39.56 | mjordan | (file remains quiet) |
| 22:40.17 | coreyfarrell | I think I started contributing around 11.2 |
| 22:40.29 | file | WHAT |
| 22:40.37 | file | I'm from pre-1.2 days |
| 22:40.57 | mjordan | file: as I established previously, those are lost in the mist of FTP |
| 22:41.29 | file | mjordan, they exist as figments in the imaginations of small children - a collective... cloud... if you will |
| 22:41.31 | coreyfarrell | mjordan: you've just given me an appreciation of svn |
| 23:51.56 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-175-118.mci.googlefiber.net) |