| 00:18.53 | *** part/#asterisk-dev kharwell (kharwell@nat/digium/x-jzbcklzayhmiivzj) |
| 05:32.37 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 07:02.50 | *** join/#asterisk-dev CELYA_ (~Thunderbi@LStLambert-657-1-90-73.w80-13.abo.wanadoo.fr) |
| 08:11.00 | *** join/#asterisk-dev tzafrir (~tzafrir@bzq-82-81-175-197.red.bezeqint.net) |
| 08:11.00 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 08:55.30 | *** join/#asterisk-dev oej (~oej@h87-96-134-129.cust.se.alltele.net) |
| 08:55.46 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 09:26.52 | *** join/#asterisk-dev oej (~oej@h87-96-134-129.cust.se.alltele.net) |
| 09:26.52 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 11:44.03 | tzafrir | If I enable TEST_FRAMEWORK, should this break the ABI to modules? |
| 11:44.44 | tzafrir | Or is it OK for me to add it to ignored directives in build_tools/make_buildops.sh ? |
| 13:14.18 | gtjoseph | @tzafrir: TEST_FRAMEWORK should NOT break ABI |
| 13:16.40 | gtjoseph | Some source file have tests embedded in them though so it may make them slightly larger. |
| 13:34.11 | tzafrir | What would be the down sides for installing test modules on a production system? |
| 13:34.39 | tzafrir | (I'm leaning towards moving them to a separate binary package, but there may be technical issues in doing that) |
| 13:34.53 | tzafrir | Apart from extra disk space usage |
| 13:43.35 | gtjoseph | The only downside I can think of is someone running them on a production system. A few of those tests do thrash tests which will push a core to 100% for up to 20 seconds |
| 14:09.33 | *** join/#asterisk-dev Marquis42 (~Marquis42@d14-69-130-192.try.wideopenwest.com) |
| 14:09.33 | *** mode/#asterisk-dev [+o Marquis42] by ChanServ |
| 14:16.25 | tzafrir | gtjoseph, but no test is run automatically (at module load time), right? |
| 14:16.34 | gtjoseph | nope |
| 14:17.02 | tzafrir | So the problem here is that it increases the attack surface |
| 14:17.11 | tzafrir | (potentially) |
| 14:17.21 | gtjoseph | correct |
| 14:19.08 | gtjoseph | oh, just thought of something else.... it will cause additional AMI events to be generated. For instance , there's an extra event generated with a contact qualifies. |
| 14:19.24 | gtjoseph | when a contact qualifies (pjsip) |
| 14:19.33 | gtjoseph | the testsuite uses it |
| 14:35.06 | *** join/#asterisk-dev kharwell (kharwell@nat/digium/x-zehmqthnewumzeml) |
| 14:35.06 | *** mode/#asterisk-dev [+o kharwell] by ChanServ |
| 15:13.51 | *** join/#asterisk-dev tzafrir (~tzafrir@bzq-82-81-175-197.red.bezeqint.net) |
| 15:13.51 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 15:19.19 | *** join/#asterisk-dev rmudgett (rmudgett@nat/digium/x-syddawcoogdhrqha) |
| 15:19.19 | *** mode/#asterisk-dev [+o rmudgett] by ChanServ |
| 15:19.58 | *** join/#asterisk-dev mjordan (mjordan@nat/digium/x-gispznfsxuqzqzvk) |
| 15:19.58 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 15:50.15 | *** join/#asterisk-dev tzafrir (~tzafrir@bzq-82-81-175-197.red.bezeqint.net) |
| 15:50.15 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 15:52.34 | *** join/#asterisk-dev jameswf (uid27319@gateway/web/irccloud.com/x-lfbcljwxygfaijee) |
| 16:34.13 | *** join/#asterisk-dev CELYA_ (~Thunderbi@2a01cb0583d31d00412823e2ddf0e8b5.ipv6.abo.wanadoo.fr) |
| 16:38.36 | *** join/#asterisk-dev tzafrir (~tzafrir@bzq-82-81-175-197.red.bezeqint.net) |
| 16:38.37 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 16:43.28 | coreyfarrell | tzafrir: TEST_FRAMEWORK does break the ABI. to ignore it you'd have to remove all references to TEST_FRAMEWORK from all headers plus ensure that test.c has stubs for all functions. doing so would break the API / probably cause compile failures in many sources. |
| 16:44.48 | coreyfarrell | gtjoseph: ^^ fyi |
| 16:45.01 | tzafrir | OK. Thanks |
| 16:45.18 | gtjoseph | really? |
| 16:45.24 | gtjoseph | where does it break the ABI? |
| 16:45.52 | coreyfarrell | if you compile a module with TEST_FRAMEWORK enabled it will not load on asterisk that was compiled with TEST_FRAMEWORK disabled (assuming the module has tests). |
| 16:46.24 | gtjoseph | oh, true |
| 16:46.32 | gtjoseph | i was looking the other way |
| 16:46.57 | gtjoseph | if you build asterisk WITH TEST_FRAMEWORK, a module built WITHOUT it should work fine. |
| 16:47.10 | rmudgett | It doesn't change existing API calls. It adds new ones. |
| 16:47.11 | coreyfarrell | in the other way it would load but the module would exclude the tests.. |
| 16:47.59 | gtjoseph | I thought tzafrir was coming from the perspective of building asterisk WITH and using extertnal modules build WITHOUT |
| 16:48.51 | coreyfarrell | gtjoseph: in any case to ignore from make_buildopts.sh it has to work in both directions. |
| 16:49.04 | tzafrir | right |
| 16:49.30 | gtjoseph | @tzafrir: sorry if i gave you bad info |
| 17:28.57 | *** join/#asterisk-dev newtonr (~newtonr@173-17-133-211.client.mchsi.com) |
| 17:28.57 | *** mode/#asterisk-dev [+o newtonr] by ChanServ |
| 17:30.20 | rmudgett | gtjoseph: I keep getting bundled pjproject compile failures on an initial build. pjsua cannot find libresample-x86_64-unknown-linux-gnu on the first make attempt |
| 17:30.38 | gtjoseph | what branch? |
| 17:31.13 | rmudgett | 13 |
| 17:31.18 | gtjoseph | current? |
| 17:31.22 | rmudgett | git |
| 17:31.24 | rmudgett | yes |
| 17:31.44 | rmudgett | seems fairly recent too |
| 17:31.53 | gtjoseph | 815f7551550908c83220196ba08742af0c745772 should have fixed that |
| 17:31.56 | rmudgett | like in the last week |
| 17:32.01 | gtjoseph | jeeeezus |
| 17:33.01 | gtjoseph | tests |
| 17:33.22 | rmudgett | I just do a 'make' No -j option |
| 17:33.41 | gtjoseph | that shouild be the best case although slow |
| 17:34.21 | rmudgett | I haven't gotten into the habit of using -j yet. :) |
| 17:38.59 | gtjoseph | rmudgett: can you try with a distclean? |
| 17:40.10 | rmudgett | I've been able to duplicate it by deleting third-party/pjproject/build.mak and the source directory |
| 17:40.36 | gtjoseph | do you do that normally? |
| 17:40.45 | rmudgett | Dislike doing distclean. It messes up my make menuselect options. |
| 17:41.08 | gtjoseph | so this is a "off-nominal" case :) |
| 17:41.08 | rmudgett | Yes I regularly delete those to force reapplication of the patches. |
| 17:41.32 | rmudgett | when I re-run configure |
| 17:42.34 | gtjoseph | which configure? asterisk or pjproject? |
| 17:42.46 | rmudgett | Also a distclean makes it forget that I have a cached pjproject source tar file |
| 17:42.58 | rmudgett | Asterisk's configure. |
| 17:44.09 | rmudgett | I use ./configure --enable-dev-mode --with-pjproject-bundled --with-externals-cache=/home/rmudgett/cache_pjproject |
| 17:44.31 | gtjoseph | can i tackle this over the weekend? |
| 17:45.06 | rmudgett | No hurry as you just have to run make again after it fails. |
| 18:25.23 | *** join/#asterisk-dev oej (~oej@h87-96-134-129.cust.se.alltele.net) |
| 18:25.23 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 18:37.51 | Kobaz | [2016-12-23 13:37:01.781] WARNING[15839]: chan_sip.c:3371 __sip_xmit: sip_xmit of 0x7f818e313d00 (len 675) to 192.168.70.8:59584 returned -2: Interrupted system call |
| 18:37.57 | Kobaz | what could be causing that? |
| 18:55.53 | *** join/#asterisk-dev sh_smith (~sh_smith@cpe-76-174-26-91.socal.res.rr.com) |
| 19:42.10 | coreyfarrell | Kobaz: looks like errno is only correctly set for UDP transport, so if you are dealing with TCP or websocket transport it's an unknown error. |
| 20:44.58 | *** join/#asterisk-dev oej (~oej@h87-96-134-129.cust.se.alltele.net) |
| 20:44.58 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 22:58.32 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 23:35.16 | *** part/#asterisk-dev kharwell (kharwell@nat/digium/x-zehmqthnewumzeml) |