| 00:13.17 | *** join/#asterisk-dev reconwireless (uid10170@gateway/web/irccloud.com/x-ptuemwgmybweudhw) |
| 01:00.48 | *** join/#asterisk-dev X-Rob (sid14615@gateway/web/irccloud.com/x-myimjlvabskuklzy) |
| 02:03.52 | *** join/#asterisk-dev tm1000 (tm1000@2600:3c01::f03c:91ff:fe70:fce6) |
| 02:52.24 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.125.148) |
| 02:53.28 | *** join/#asterisk-dev tonyclewis (uid6025@gateway/web/irccloud.com/x-bmnzhhxcaujpuoxr) |
| 03:46.25 | tonyclewis | X-Rob: did you ask file about the core dumps we get when doing a core reload |
| 04:00.01 | tm1000 | tonyclewis: we havent filed anything with asterisk yet. trying to narrow down first |
| 04:01.14 | *** join/#asterisk-dev WIMPy (~wimpy@e183095026.adsl.alicedsl.de) |
| 04:17.34 | X-Rob | tm1000: I kinda disagree with the 'hide chan_sip/pjsip trunks when they're not enabled' |
| 04:18.00 | tm1000 | X-Rob: ok. thats an easy fix though :-) |
| 04:30.05 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.125.148) |
| 04:54.20 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.125.139) |
| 05:06.14 | *** join/#asterisk-dev GameGamer43 (sid5533@gateway/web/irccloud.com/x-wfdtufevykcddcwv) |
| 05:18.02 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.125.230) |
| 05:23.48 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.125.230) |
| 05:54.13 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.125.207) |
| 06:54.37 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.125.210) |
| 07:07.20 | *** join/#asterisk-dev reconwireless (uid10170@gateway/web/irccloud.com/x-mghqzpblpgxujlld) |
| 07:24.44 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.125.174) |
| 09:16.06 | *** join/#asterisk-dev schmidts (~schmidts@pat65.net.uta.at) |
| 09:16.08 | schmidts | good morning |
| 09:33.53 | oej | Morning folks! |
| 09:34.05 | oej | Stefan - thank you for your report. Made me very happy. |
| 09:34.17 | oej | We need to figure out the missing parts quickly. |
| 09:41.42 | schmidts | oej there is allready a patch on reviewboard for this issue and i am allready trying it |
| 09:42.04 | schmidts | but i hit another problem with this, which looks even worse if this comes from pjsip itself not from this patch |
| 09:43.00 | schmidts | i have tried to reproduce it, without this patch just with normal notifys and response their but then this doesnt happens |
| 11:19.59 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.125.226) |
| 13:40.40 | *** join/#asterisk-dev roderickm (~roderickm@67.63.143.254) |
| 14:23.16 | *** join/#asterisk-dev danjenkins (~dan@cpc3-folk2-2-0-cust34.1-2.cable.virginm.net) |
| 14:24.44 | danjenkins | bopnjour all, just trying to install ast 12, I've uninstalled pjproject, pulled it down from github and reinstalled it but I'm getting an error when doing make in ast "‘pjsip_transaction’ has no member named ‘mutex’" - wondered if someone could tell me what I've done wrong? |
| 14:26.11 | file | did you re-run configure? |
| 14:26.19 | danjenkins | i thought i had |
| 14:26.46 | danjenkins | ah maybe not after uninstalling pjproject completely |
| 14:27.28 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 14:27.28 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:27.35 | danjenkins | that was it |
| 14:27.40 | danjenkins | what would i do without you file |
| 14:27.47 | file | not have cookies! |
| 14:28.37 | danjenkins | indeed! |
| 14:30.51 | file | danjenkins, how is day #0 treating you? |
| 14:31.23 | danjenkins | day #0 is monday or are you implying today is the day before day 1 |
| 14:31.37 | file | today is the day before day 1 |
| 14:37.10 | blitzrage | danjenkins: run/. |
| 14:38.21 | danjenkins | @file day #0 is going well, trying to sort out my ast 12 install finally and need to tidy my office in prep for monday |
| 14:38.35 | danjenkins | thanks !blitzrage but file told me how stupid i was being |
| 14:38.41 | file | torch it! |
| 14:38.44 | blitzrage | ok good |
| 14:44.25 | *** join/#asterisk-dev vlad_sta_ (~vlad_star@109.188.125.226) |
| 14:44.53 | *** join/#asterisk-dev leifmadsen (~free-mads@asterisk/documenteur-extraordinaire/blitzrage) |
| 14:44.53 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 14:48.14 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.125.226) |
| 15:05.21 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@64.235.216.2) |
| 15:11.07 | *** join/#asterisk-dev elred_ (~elred_@unaffiliated/elred-/x-5010831) |
| 15:29.22 | *** join/#asterisk-dev pc-m (~pcm@modemcable094.94-70-69.static.videotron.ca) |
| 15:29.47 | *** join/#asterisk-dev pc-m (~Thunderbi@modemcable094.94-70-69.static.videotron.ca) |
| 15:32.18 | *** join/#asterisk-dev scouture (~scouture@unaffiliated/scouture) |
| 15:39.07 | *** join/#asterisk-dev kharwell (~kharwell@nat/digium/x-rjpctyumdfgrgmee) |
| 15:40.34 | *** join/#asterisk-dev mjordan (~matt@nat/digium/x-gxvzgrwwtnshkwpw) |
| 15:40.34 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 15:55.29 | *** join/#asterisk-dev elred_ (~elred_@unaffiliated/elred-/x-5010831) |
| 15:59.10 | *** join/#asterisk-dev gtjoseph (~gtjoseph@unaffiliated/gtj) |
| 16:02.17 | *** join/#asterisk-dev danjenkins (~dan@cpc3-folk2-2-0-cust34.1-2.cable.virginm.net) |
| 16:06.36 | *** join/#asterisk-dev otherwiseguy (~otherwise@99-46-2-71.lightspeed.mssnks.sbcglobal.net) |
| 16:10.58 | *** join/#asterisk-dev ipengineer (~zconkle@static-71-252-134-63.dllstx.fios.verizon.net) |
| 16:17.55 | *** join/#asterisk-dev angler (~angler@pdpc/sponsor/digium/angler) |
| 16:18.41 | gtjoseph | Any known issues with turning PJ_HAS_IPV6 on in pjproject? It's default is off. |
| 16:18.49 | file | nope |
| 16:18.51 | file | works fine |
| 16:18.58 | gtjoseph | cool |
| 16:50.46 | *** part/#asterisk-dev volga629 (~bendersky@CPE085b0e07d3f2-CM7cb21b15b251.cpe.net.cable.rogers.com) |
| 16:55.34 | tm1000 | file: do you have any issues with "core reload" |
| 16:56.07 | file | personally? no, I find it fine |
| 16:56.26 | tm1000 | file: it core dumps our versions of asterisk. |
| 16:56.34 | tm1000 | sooooo we are going to debug and get back to you :-) |
| 16:56.35 | file | I haven't seen any reported issues and I haven't experienced any problems |
| 16:56.54 | mjordan | tm1000: haven't hit it running it locally |
| 16:57.05 | mjordan | tm1000: as soon as you snag a backtrace let me know |
| 16:57.22 | tm1000 | we have some backtraces but dont wanna just waste your time :-)…..yet |
| 16:57.37 | mjordan | well, it *shouldn't* crash |
| 16:57.56 | mjordan | if you want to pb one, we can at least get an idea of what is blowing up |
| 16:59.38 | tm1000 | mjordan: http://pastebin.com/1Ln4Jbh0 |
| 16:59.43 | tm1000 | file: ^ |
| 17:00.08 | file | oh |
| 17:00.22 | file | MY FAULT |
| 17:00.27 | Qwell | file: BAD FILE |
| 17:00.37 | mjordan | strerrno: eagain? |
| 17:01.07 | file | easy to fix! |
| 17:01.14 | file | mjordan, pjlib usage in non-pj thread |
| 17:01.22 | mjordan | indeed |
| 17:01.24 | tm1000 | file: oh dont worry. I have more core dumps :-p |
| 17:01.26 | Qwell | file: doesn't it warn on that or something? |
| 17:01.36 | file | it asserts. |
| 17:01.38 | mjordan | and I'm not running outbound registration, so that would be why I didn't hit it |
| 17:01.52 | Qwell | oh, right, pj_thread_this |
| 17:02.05 | tm1000 | mjordan: file if you want me to file this for tracking purposes let me know |
| 17:02.20 | mjordan | tm1000: unless you already have half a ticket filed, don't worry bout it |
| 17:02.28 | tm1000 | k |
| 17:02.38 | tm1000 | well I will send you my other fun one in a minute, although they might be related |
| 17:02.57 | tm1000 | (the other fun one is when an endpoint attempts to register…core dump) |
| 17:03.05 | mjordan | hm |
| 17:03.09 | file | that's... odd |
| 17:03.09 | tm1000 | but I need to be able to replicate it better |
| 17:03.11 | mjordan | that doesn't sound necessarily related |
| 17:03.17 | file | yeah, doubt it |
| 17:03.21 | tm1000 | let me see if it happens on another system first |
| 17:03.23 | mjordan | if you have a proper backtrace, I'd still want to see |
| 17:03.37 | tm1000 | mjordan: np let me just get the backtrace from that as well |
| 17:04.26 | *** join/#asterisk-dev danjenkins (~dan@cpc3-folk2-2-0-cust34.1-2.cable.virginm.net) |
| 17:05.45 | *** join/#asterisk-dev otherwiseguy (~otherwise@me42836d0.tmodns.net) |
| 17:07.03 | tm1000 | mjordan: first here is whats happening (now let me get the backtrace) |
| 17:07.03 | tm1000 | http://pastebin.com/39tyK2CF |
| 17:08.50 | tm1000 | associated core dump for endpoint registering (locally not trunks): http://pastebin.com/paz35TbC |
| 17:08.55 | tm1000 | mjordan: file ^ |
| 17:09.01 | file | uno sec |
| 17:10.04 | file | oh my |
| 17:10.24 | mjordan | fark |
| 17:10.41 | mjordan | that alloca there might not have been a good idea |
| 17:11.22 | tm1000 | mjordan: file we were contemplating if its "32b v 64b" |
| 17:11.34 | tm1000 | as a 64bit system doesnt have that issue, only this 32b one |
| 17:11.53 | file | reload should be good |
| 17:12.05 | mjordan | it's blowing the stack - most likely because the data being pushed into the stack allocated string is larger than the space allocated, and it can't expand the size of the string any larger |
| 17:13.19 | mjordan | oh |
| 17:13.26 | mjordan | nonce is actually allocated on a stack frame above it |
| 17:14.53 | tm1000 | mjordan: want me to file a bug on it? |
| 17:15.25 | mjordan | well, there's an easy fix. I'm not positive it's the right one. We can just heap allocate the ast_str objects which will avoid the stack smashing. |
| 17:16.01 | tm1000 | yay! |
| 17:16.02 | tm1000 | dances |
| 17:16.31 | mjordan | file: what do you think? |
| 17:16.42 | file | I'm fixing a few more URI validation cases |
| 17:16.53 | file | but that is what came to mind |
| 17:17.44 | putnopvut | I'm having a look too. |
| 17:17.55 | putnopvut | Seems unlikely though, unless the realm name is really large. |
| 17:20.16 | putnopvut | Oh wait, well that's silly. That one frame uses a 256 byte allocation, but the one previous uses a 64 byte allocation >_< |
| 17:20.18 | mjordan | I don't think its the stack allocated one in build_nonce |
| 17:20.23 | mjordan | yup :-) |
| 17:20.34 | mjordan | but it actually blew up along the challenge path |
| 17:20.36 | putnopvut | So yeah, not too surprising to blow that. |
| 17:20.37 | mjordan | which does use a 256 byte allocation |
| 17:22.00 | mjordan | tm1000: crappy patch to try - http://pastebin.com/8Su7jKbb |
| 17:22.51 | tm1000 | file: is your reload patch in svn? |
| 17:22.56 | file | yes |
| 17:23.20 | mjordan | the error handling of things calling challenge should probably change |
| 17:23.22 | putnopvut | Actually, the 64 byte allocation isn't all that crazy. The 256 byte string is then hashed to md5 (lowering it to 32 bytes). Then a timestamp and / is prepended, meaning that 64 should be way mroe than enough for the calculated nonce. |
| 17:25.22 | mjordan | putnopvut: we do have a lot of allocations on the stack leading up to that point however |
| 17:25.30 | putnopvut | yeah, I was wondering about that. |
| 17:25.41 | mjordan | digest_check_auth, for example, does two that are potentially large |
| 17:25.54 | mjordan | the right patch might be to make those heap allocated |
| 17:25.55 | putnopvut | I guess the fact that there is a 256 byte string allocation in challenge() isn't helping matters. |
| 17:30.26 | tm1000 | mjordan: compiling now |
| 17:37.45 | blitzrage | mjordan: so I've heard you're adding HEPv3 support to chan_pjsip |
| 17:37.51 | blitzrage | I approve of this. |
| 17:37.59 | mjordan | blitzrage: orly? |
| 17:38.07 | mjordan | you guys use it often? |
| 17:38.14 | blitzrage | not yet, but we plan to at some point |
| 17:38.31 | blitzrage | would love to be able to backport it to chan_sip since we're on 1.8 still :\ |
| 17:38.40 | blitzrage | hoping to add some Asterisk 12/13 in the next 12-16 months |
| 17:38.43 | mjordan | blitzrage: that's actually the patch that alexandr has |
| 17:38.48 | blitzrage | orly |
| 17:38.50 | mjordan | rly |
| 17:39.00 | mjordan | what all do you plan on using it for? SIP call flows, or other stuff too? |
| 17:39.40 | blitzrage | mjordan: ya, basically ways of being able to monitor all our traffic, and be able to pull out issues with certain calls and analyze them after the fact |
| 17:39.48 | blitzrage | mostly for support stuff on our infrastructure side |
| 17:39.52 | mjordan | is RTCP interesting? |
| 17:39.55 | blitzrage | hells ya |
| 17:40.12 | blitzrage | would love to be able to monitor status of carriers and be able to auto switch based on data |
| 17:40.22 | blitzrage | it's been asked of me several times in the last year |
| 17:40.26 | mjordan | k |
| 17:40.31 | blitzrage | I just have to keep telling people it's not something we can rely on in Asterisk |
| 17:40.40 | mjordan | that's the next module I was going to write |
| 17:40.46 | blitzrage | I would approve of this as well. |
| 17:40.46 | mjordan | haven't tested *anything* yet |
| 17:40.50 | blitzrage | heh I hear ya :) |
| 17:40.57 | blitzrage | just write it all and then come back and clean it up later :) |
| 17:40.58 | mjordan | but got it to compile last night while watching Alabama get creamed |
| 17:41.07 | mjordan | blitzrage: the open source way |
| 17:41.13 | blitzrage | sounds right |
| 17:41.45 | blitzrage | can't wait till we get a couple more people in a tier 2 position so I can just focus on developing new features and deployments |
| 17:41.52 | blitzrage | right now I do far too much tier 2 support |
| 17:41.54 | mjordan | If you can think of use cases outside of RTCP information and SIP packets, let me know. |
| 17:41.57 | blitzrage | well, my entire team does |
| 17:42.05 | mjordan | "customers" |
| 17:42.31 | blitzrage | mjordan: ya, basically ability to monitor calls in some live method is useful as well. We haven't enabled it yet, but we plan to hook into SNMP at some point. |
| 17:42.34 | mjordan | http://demotivators.despair.com/demotivational/apathydemotivator.jpg |
| 17:42.47 | blitzrage | nice |
| 17:45.01 | tm1000 | mjordan: file reload is fixed. endpoint registration is not |
| 17:45.11 | mjordan | drat |
| 17:45.19 | mjordan | does the backtrace look the same? |
| 17:45.26 | tm1000 | mjordan: want another backtrace |
| 17:45.36 | mjordan | sure, just to see if I made it angrier |
| 17:46.52 | tm1000 | mjordan: one sec I didnt enable backtraces ….. ugh myself |
| 17:46.54 | *** join/#asterisk-dev micols (~gnu@shell1.rlogin.dk) |
| 17:49.09 | danjenkins | could someone check what my asterisk jira login is? know i def had one but might have been setup in my previous employers email address |
| 17:49.44 | mjordan | danjenkins: yup, still your old e-mail addr |
| 17:49.48 | mjordan | which one would you prefer? |
| 17:50.05 | danjenkins | dan@nimbleape.com would be awesome |
| 17:50.23 | mjordan | updated! |
| 17:50.29 | danjenkins | thanks! |
| 17:52.26 | tm1000 | mjordan: http://pastebin.com/ggrqUVeZ |
| 17:52.47 | mjordan | wow. That's exactly the same :-( |
| 17:53.38 | mjordan | what's the distro? CentOS 6 32-bit? |
| 17:53.44 | tm1000 | mjordan: yes |
| 17:56.56 | tm1000 | mjordan: if you want access to that machine just say the word |
| 17:57.40 | mjordan | Do you still have the core file? |
| 17:58.40 | tm1000 | yes |
| 17:58.44 | mjordan | sure :-) |
| 17:59.05 | tm1000 | mjordan: PM GameGamer43 |
| 18:06.07 | danjenkins | just filed an issue https://issues.asterisk.org/jira/browse/ASTERISK-23092 - not sure if its just me or not but there you go |
| 18:10.16 | *** join/#asterisk-dev danjenkins (~dan@cpc3-folk2-2-0-cust34.1-2.cable.virginm.net) |
| 18:10.57 | danjenkins | oops irc crashed |
| 18:13.18 | danjenkins | I'm off |
| 18:14.52 | *** join/#asterisk-dev sruffell (~sruffell@asterisk/the-kernel-guy/sruffell) |
| 18:14.52 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 18:34.47 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 18:34.47 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 18:48.54 | mjordan | well, this is definitely odd. |
| 18:50.17 | putnopvut | mjordan: let me guess, a number not divisible by 2? |
| 18:50.22 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 18:51.03 | [TK]D-Fender | Every number is divisible by two |
| 18:51.13 | [TK]D-Fender | #downwithintegerscum |
| 18:51.16 | mjordan | putnopvut: I think I have a suspicion where its going wrong |
| 18:51.30 | mjordan | so, nothing is all that large on the stack |
| 18:51.32 | putnopvut | mjordan: what are we talking about exactly? I just joined this party not that long ago. |
| 18:51.36 | putnopvut | Oh, the stack thing |
| 18:51.41 | putnopvut | Yeah, so what's up? |
| 18:51.46 | mjordan | res_pjsip_digest_authenticator is, however, allocating an array of ast_sip_auth pointers |
| 18:51.54 | mjordan | it then tries to set string fields in there. |
| 18:52.02 | mjordan | it doesn't init the string field memory pool (which is bad) |
| 18:52.27 | mjordan | but since it doesn't init the string field memory pool, I wonder if setting the realm to "asterisk" is blowing it |
| 18:53.01 | putnopvut | mjordan: it should just be getting already allocated ast_sip_auth structures, that already have the string fields init'd |
| 18:53.25 | tzafrir | I can't seem to download attachments from the JIRA with wget. |
| 18:53.26 | mjordan | goes to look at the endpoint |
| 18:53.29 | putnopvut | The array is alloca'd in the digest_check_auth, but it should be retrieving ast_sip_auth structures already allocated by sorcery |
| 18:53.43 | tzafrir | (GET works just fine) |
| 18:57.07 | mjordan | It does look valid. I'm honestly not sure what is smashing the stack here. None of the stack allocations are large, but something is sure screwed up. |
| 18:58.02 | mjordan | tzafrir: I'm not sure, I haven't tried that in awhile. It's possible it's trying to authenticate |
| 18:58.35 | tzafrir | mjordan, isn't surely not that: 'GET' works just fine |
| 18:58.42 | tzafrir | Maybe it's an issue of user agent |
| 18:59.50 | tzafrir | no |
| 19:00.10 | tzafrir | gnutls issue? |
| 19:01.46 | tzafrir | gnutls-cli is happy with it |
| 19:05.54 | tzafrir | also: review-board doesn't like me pasting test with the middle button (browser: iceweasel) |
| 19:06.18 | tzafrir | When I do that: the text is not accepted. Not sure why. I haven't noticed this before |
| 19:06.36 | mjordan | we haven't updated review board |
| 19:08.45 | tzafrir | Though adding even a space in the end or the beggining seems to work around that |
| 19:18.16 | tzafrir | On a related issue: any progress with merging the pjproject patches? I thought this would happen by the time Asterisk 12 is released |
| 19:23.31 | pabelanger | tzafrir, thanks for fixing that annoying warning |
| 19:31.40 | mjordan | a lot of patches have been merged. I'm not sure how many are left to go, but Teluu is working through them. |
| 19:32.01 | file | I don't think we have any active ones pending |
| 19:35.31 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.125.226) |
| 19:42.32 | *** join/#asterisk-dev ipengineer (~zconkle@static-71-252-134-63.dllstx.fios.verizon.net) |
| 19:45.23 | *** join/#asterisk-dev hexanol (~bibi@modemcable094.94-70-69.static.videotron.ca) |
| 19:50.14 | *** join/#asterisk-dev ipengineer (~zconkle@static-71-252-134-63.dllstx.fios.verizon.net) |
| 19:53.39 | mjordan | putnopvut: so. This error should occur only when something writes passed the bounds of something allocated on the stack |
| 19:54.05 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.125.226) |
| 19:54.25 | mjordan | and the only thing that seems to be a candidate for that is hash |
| 19:59.44 | putnopvut | mjordan: yeaaahhh |
| 20:00.15 | putnopvut | So md5 hashes are 32 bytes, so probably should size that array to 33 to account for the null byte. |
| 20:04.42 | *** join/#asterisk-dev ipengineer (~zconkle@static-71-252-134-63.dllstx.fios.verizon.net) |
| 20:15.20 | *** part/#asterisk-dev hexanol (~bibi@modemcable094.94-70-69.static.videotron.ca) |
| 20:34.51 | mjordan | That's about the only thing I can think of. |
| 20:38.59 | *** join/#asterisk-dev danjenkins (~dan@cpc3-folk2-2-0-cust34.1-2.cable.virginm.net) |
| 20:57.47 | mjordan | putnopvut: that was it |
| 20:58.12 | putnopvut | oh joyous joy |
| 20:58.35 | mjordan | now we get to look at what else calls ast_md5_hash :-\ |
| 21:02.32 | mjordan | tm1000: I'll get that cleaned up and committed in a bit |
| 21:04.08 | putnopvut | mjordan: looks like the only other bad place is in test_utils.c |
| 21:05.51 | mjordan | yup, compiling now |
| 21:06.16 | mjordan | tm1000: thanks for testing it out :-) |
| 21:20.32 | X-Rob | newtonr: Well. That's just strange. |
| 21:21.26 | newtonr | X-Rob, yes, it is quite confusing |
| 21:21.56 | file | it's not weird if you know what the code is doing |
| 21:22.20 | file | dislikes the code that does it |
| 21:22.57 | X-Rob | hands file some kerosine and matches. |
| 21:24.11 | file | I see the difference though |
| 21:25.10 | file | chan_pjsip uses the requested formats, chan_sip appends the requested formats |
| 21:26.17 | X-Rob | was reading scrollback |
| 21:26.38 | X-Rob | mjordan: putnopvut : Do I see you say you've found the crash? (This is tm1000's crash, right?) |
| 21:26.50 | file | both have been fixed |
| 21:26.55 | file | the core reload crash and the registration crash |
| 21:27.00 | X-Rob | <3 |
| 21:27.06 | X-Rob | Man. I love open source. |
| 21:27.14 | X-Rob | This stuff is far too much fun. |
| 21:34.15 | pabelanger | Free development |
| 21:38.04 | *** join/#asterisk-dev otherwiseguy (~otherwise@2605:a601:437:ce01:3e97:eff:fe14:ef66) |
| 21:38.55 | tm1000 | X-Rob: newtonr |
| 21:39.01 | tm1000 | Rusty Meet Rob |
| 21:39.09 | X-Rob | tm1000: we already did 8) |
| 21:40.03 | newtonr | HI! I like running obstacle courses, staring at Asterisk logs and long walks on the beach! |
| 21:42.44 | *** join/#asterisk-dev leedm777 (~leedm777@nat/digium/x-jfnsuqjcfzycuygu) |
| 21:46.00 | X-Rob | newtonr: Hi. I'm a Leo, I believe in spiritual haunting (of code), and my favourite cake is Pi. |
| 22:33.12 | mjordan | so here's a completely different question: ConfBridge with a single marked user and 3 waitmarked users. Everyone is hanging out. The marked user leaves. Do the waitmarked users continue to enjoy the conference experience, or do we put them back into an inactive/moh state? |
| 22:35.12 | newtonr | they should continue to enjoy unless end_marked is set for them |
| 22:36.28 | newtonr | i'm saying what i think should happen. I can't remember how it currently behaves in that scenario. |
| 22:40.21 | tm1000 | mjordan: file thanks for all of the help today. Asterisk12 works on 32bit and 64 bit with the mjordan patch |
| 22:40.28 | file | nifty |
| 22:40.44 | tm1000 | file: not sure if matt talked to you about the hash[33] patch? |
| 22:40.50 | file | I saw |
| 22:43.56 | mjordan | not a problem :-) |
| 22:44.13 | file | I SEE ALL |
| 22:44.51 | mjordan | newtonr: I'm willing to take your opinion as gospel |
| 22:49.41 | mjordan | newtonr: actually, there was a bug in the code: but the goal was to switch to INACTIVE |
| 22:49.54 | mjordan | and since I don't want to add *yet another state* to ConfBridge, I may go with that |
| 22:51.12 | newtonr | fair enough |
| 22:55.08 | newtonr | mjordan, If things happen when the last marked user leaves, then we should probably modify the documenation too; for wait_marked it only says "Sets if the user must wait for a marked user to enter before joining the conference. Off by default." |
| 22:55.38 | newtonr | That is the sample file. Not sure about other confbridge or option documentation |
| 22:56.15 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 22:59.25 | WIMPy | Couldn't that flag just be cleared once a maked user enters? |
| 23:00.00 | mjordan | WIMPy: you'd think so |
| 23:00.07 | mjordan | but that actually was messier |
| 23:00.28 | mjordan | early ConfBridge was a train wreck of odd edge corners due to interactions between marked, waitmarked, and users with no marked flags |
| 23:00.42 | mjordan | for example: you have two waitmarked users (inactive) and a user with no flags joins |
| 23:00.59 | WIMPy | I can imagine that kind of stuff getting messy. |
| 23:01.25 | mjordan | we've got it pretty well in hand, there's just a goofy race condition when you have marked users and waitmarked users that shouldn't be there |
| 23:10.38 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.125.226) |
| 23:14.03 | tonyclewis | and we have liftoff complete ISO of FreePBX 12 and Asteirsk 12 |
| 23:14.06 | tonyclewis | compiled and installing |
| 23:18.03 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.125.226) |
| 23:22.27 | newtonr | tonyclewis, awesome! |
| 23:40.56 | *** part/#asterisk-dev kharwell (~kharwell@nat/digium/x-rjpctyumdfgrgmee) |
