| 00:25.11 | *** join/#asterisk-dev spatel (~spatel@pool-98-118-124-216.bstnma.fios.verizon.net) |
| 05:33.54 | *** join/#asterisk-dev jkroon (~jkroon@165.16.204.106) |
| 06:10.04 | *** join/#asterisk-dev CELYA_ (~Thunderbi@2a01:e0a:d:1111:b749:768d:ed77:7c9a) |
| 06:44.54 | *** join/#asterisk-dev jkroon (~jkroon@165.16.203.53) |
| 06:48.08 | *** join/#asterisk-dev jkroon (~jkroon@165.16.203.53) |
| 06:49.00 | *** join/#asterisk-dev pchero_work (~pchero@dhcp-077-249-058-090.chello.nl) |
| 06:51.19 | *** join/#asterisk-dev jkroon (~jkroon@165.16.203.53) |
| 07:04.02 | *** join/#asterisk-dev hehol (~hehol@gatekeeper.loca.net) |
| 08:41.30 | *** join/#asterisk-dev pchero_work (~pchero@51.247.195.35.bc.googleusercontent.com) |
| 10:29.27 | *** join/#asterisk-dev pchero_work (~pchero@87.213.247.82) |
| 11:20.26 | *** join/#asterisk-dev bford (uid283514@gateway/web/irccloud.com/x-pcfbqxyujjdntqdh) |
| 11:20.26 | *** mode/#asterisk-dev [+o bford] by ChanServ |
| 11:48.47 | seanbright | i don't know why... but i feel like ASTERISK-28576 is somehow related to https://gerrit.asterisk.org/c/asterisk/+/11467 |
| 11:49.40 | seanbright | i have 0 evidence to support that theory |
| 11:50.57 | file | did it work in 16.5.0? |
| 11:51.02 | file | (and not crash) |
| 11:51.21 | file | I ask as 16.5.0 and 16.6.0 are equivalent for res_rtp_asterisk.c |
| 11:53.34 | *** join/#asterisk-dev pc-m (~pcm@ipagstaticip-cf899324-841a-2614-948a-7dbf21746b29.sdsl.bell.ca) |
| 12:14.07 | seanbright | i do have a similar crash i think |
| 12:14.18 | seanbright | let me see if i can find it |
| 12:16.53 | seanbright | https://pastebin.com/raw/ZGp5BRTC |
| 12:17.10 | seanbright | that's with 1158411f53fc0b4d22511c2527850cad457b9c4a as HEAD |
| 12:17.31 | seanbright | OpenSSL 1.0.2g |
| 12:17.37 | file | nods |
| 12:19.00 | seanbright | so after 16.5 unfortunately |
| 12:19.55 | seanbright | and it doesn't crash frequently for me |
| 12:20.24 | seanbright | two crashes in the last 3 months across 8 servers. not sure what the other crash was. |
| 12:20.39 | file | interestin' |
| 12:21.31 | seanbright | same crash |
| 12:21.33 | seanbright | both were the same |
| 12:23.50 | seanbright | so in 1.0.2g, there are a few places there OPENSSL_assert is called |
| 12:24.08 | seanbright | OPENSSL_assert(s->d1->mtu >= dtls1_min_mtu(s)); |
| 12:24.50 | seanbright | hmm, let me install debug symbols |
| 12:27.12 | seanbright | yeah, so one of the asserts in dtls1_do_write is firing |
| 12:29.40 | seanbright | there are 3 total asserts in that function in my version |
| 12:30.54 | seanbright | https://github.com/openssl/openssl/blob/OpenSSL_1_0_2g/ssl/d1_both.c#L264-L477 |
| 12:31.36 | seanbright | it's not an assert in newer revisions |
| 12:31.37 | file | do the debug symbols yield enough to let you examine the values and narrow it down? |
| 12:32.05 | seanbright | no, i just get function names |
| 12:34.50 | seanbright | the debug symbols are for a later revision of the library |
| 12:35.02 | seanbright | but i don't know that it really matters |
| 12:35.34 | *** join/#asterisk-dev CELYA_ (~Thunderbi@92.184.99.51) |
| 12:35.39 | seanbright | it's effectively the same backtrace as OP's, other than OpenSSLDie being visible |
| 12:36.00 | seanbright | sorry for being so verbose |
| 12:43.02 | *** join/#asterisk-dev CELYA_ (~Thunderbi@92.184.99.51) |
| 12:44.36 | seanbright | i wonder if i could examine the memory directly... but that might be beyond my skillz |
| 12:52.04 | *** join/#asterisk-dev jkroon (~jkroon@165.16.204.106) |
| 13:02.49 | *** join/#asterisk-dev sduthil (~sduthil@ipagstaticip-cf899324-841a-2614-948a-7dbf21746b29.sdsl.bell.ca) |
| 13:04.22 | *** join/#asterisk-dev CELYA_1 (~Thunderbi@185.122.103.46) |
| 13:05.41 | *** join/#asterisk-dev CELYA_ (~Thunderbi@185.122.103.46) |
| 13:13.58 | *** join/#asterisk-dev pchero_work (~pchero@87.213.247.82) |
| 14:07.56 | *** join/#asterisk-dev kharwell (uid358942@gateway/web/irccloud.com/x-pjafzxuyosoticqn) |
| 14:07.56 | *** mode/#asterisk-dev [+o kharwell] by ChanServ |
| 14:16.09 | seanbright | so i pulled the ubuntu source package for openssl 1.0.1f |
| 14:16.16 | seanbright | err, brb |
| 14:17.37 | file | OpenSSL versioning is a fun thing |
| 14:19.06 | *** join/#asterisk-dev CELYA_ (~Thunderbi@2a01:e0a:d:1111:b749:768d:ed77:7c9a) |
| 14:19.56 | seanbright | ok, sorry |
| 14:20.12 | seanbright | this is weird |
| 14:20.52 | seanbright | let me get it right before i say something dumb |
| 14:23.51 | seanbright | the difficulty i am facing is that i am running some painfully old versions of ubuntu |
| 14:24.00 | seanbright | but using the xenial openssl packages |
| 14:24.18 | seanbright | and 'apt-get source' is pulling the trusty sources |
| 14:25.07 | file | that would certainly give a headache |
| 14:25.55 | file | I'm kind of surprised you were able to get the xenial packages installed with everything happy |
| 14:31.12 | seanbright | it's "documented" somewhere |
| 14:31.29 | file | ah, fancy |
| 14:31.35 | seanbright | not officially |
| 14:31.56 | seanbright | anyway, finally got the sources and i am wrong |
| 14:32.08 | seanbright | i thought the new version replaced the assertion, but it does not |
| 14:32.22 | seanbright | at least the latest xenial version |
| 15:16.09 | *** join/#asterisk-dev eliel_ (~eliel@190.210.22.142) |
| 15:22.00 | *** join/#asterisk-dev eliel_ (~eliel@190.210.22.142) |
| 16:43.27 | *** join/#asterisk-dev hehol (~Adium@ip-176-199-151-97.hsi06.unitymediagroup.de) |
| 17:32.43 | seanbright | so maybe it only affects 1.0.2+ of openssl? |
| 17:32.49 | seanbright | if that isn't a complete red herring |
| 17:36.12 | *** join/#asterisk-dev infobot (ibot@c-174-52-60-165.hsd1.ut.comcast.net) |
| 17:36.12 | *** 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 |