| 00:27.26 | *** join/#asterisk-dev Sjors (~sgielen@foo.kassala.de) |
| 00:44.55 | *** join/#asterisk-dev Sjors (~sgielen@foo.kassala.de) |
| 00:54.04 | *** join/#asterisk-dev Sjors (~sgielen@foo.kassala.de) |
| 01:19.03 | *** join/#asterisk-dev Sjors (~sgielen@foo.kassala.de) |
| 01:40.04 | *** join/#asterisk-dev dms1 (~dms1@c-50-168-154-89.hsd1.sc.comcast.net) |
| 01:49.02 | *** join/#asterisk-dev Sjors (~sgielen@foo.kassala.de) |
| 03:15.55 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-175-118.mci.googlefiber.net) |
| 03:21.59 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-175-118.mci.googlefiber.net) |
| 03:51.42 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-175-118.mci.googlefiber.net) |
| 03:58.40 | *** join/#asterisk-dev gtjoseph (~gtj0@unaffiliated/gtj) |
| 04:09.24 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-175-118.mci.googlefiber.net) |
| 04:14.34 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-175-118.mci.googlefiber.net) |
| 04:20.43 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-175-118.mci.googlefiber.net) |
| 04:27.50 | *** join/#asterisk-dev drmessano (~nonya@pdpc/supporter/active/drmessano) |
| 04:43.31 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-175-118.mci.googlefiber.net) |
| 05:07.54 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-175-118.mci.googlefiber.net) |
| 05:28.56 | *** join/#asterisk-dev Sjors (~sgielen@foo.kassala.de) |
| 06:07.18 | *** join/#asterisk-dev bulkorok (~Benjamin@85.183.61.47) |
| 09:52.47 | *** join/#asterisk-dev infobot (ibot@rikers.org) |
| 09:52.47 | *** 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-beta2 now released. Do the (beta) release dance! -=- Follow on Twitter at @AsteriskDev |
| 09:54.19 | *** join/#asterisk-dev Sjors (~sgielen@foo.kassala.de) |
| 12:21.06 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@216-191-106-163.dedicated.allstream.net) |
| 12:38.16 | *** join/#asterisk-dev Sjors (~sgielen@foo.kassala.de) |
| 13:15.12 | jsmith | Morning all... |
| 13:16.03 | leifmadsen | jsmith: ohai |
| 13:16.30 | jsmith | worked things out so that he can be in Vegas for the Developers conference on Tuesday |
| 13:16.40 | jsmith | Unfortunately, I won't be able to stick around for the rest of the show :-/ |
| 13:17.00 | leifmadsen | sadness, but at least I get to hang out with you for a day! |
| 13:17.21 | jsmith | Aye, and a good day at that :-) |
| 13:17.33 | jsmith | leifmadsen: I'll even buy you lunch that day :-) |
| 13:17.55 | leifmadsen | lol |
| 13:17.56 | leifmadsen | true story |
| 13:18.08 | leifmadsen | I love buying lunch with company money :D |
| 13:24.48 | *** join/#asterisk-dev newtonr (~newtonr@nat/digium/x-yuydmifprzwzpnuo) |
| 13:24.48 | *** mode/#asterisk-dev [+o newtonr] by ChanServ |
| 13:42.12 | *** join/#asterisk-dev drmessano (~nonya@pdpc/supporter/active/drmessano) |
| 13:54.53 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-bmpbsgoaumctxvaz) |
| 13:54.53 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 14:04.06 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 14:04.06 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:05.44 | *** join/#asterisk-dev drmessano (~nonya@pdpc/supporter/active/drmessano) |
| 14:13.42 | mjordan | gtjoseph: ping |
| 14:14.00 | file | gtjoseph, ping (over a redundant link) |
| 14:14.13 | leifmadsen | mitm pong |
| 14:14.18 | mjordan | egads |
| 14:14.25 | malcolmd | leifmadsen: hah |
| 14:14.28 | leifmadsen | gtjoseph: ping |
| 14:14.52 | leifmadsen | need to make sure his pong response goes to me |
| 14:14.57 | Marquis42 | lol |
| 14:15.13 | file | I feel like I'm at SIPit again |
| 14:15.15 | Marquis42 | Ping really needs some authentication mechanisms. :D |
| 14:15.22 | file | has flashbacks to forking multiparty |
| 14:15.26 | Marquis42 | haha |
| 14:19.45 | putnopvut | Is "forking" a minced oath? |
| 14:22.20 | *** join/#asterisk-dev kharwell (~kharwell@nat/digium/x-sammlyqesmyxlbcw) |
| 14:23.10 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-175-118.mci.googlefiber.net) |
| 14:26.44 | Qwell | putnopvut: it's France, don't judge |
| 14:29.45 | file | one more timeeeee |
| 14:33.17 | gtjoseph | mjordan, file, leifmadsen: pong, pong, pong |
| 14:33.39 | gtjoseph | what did I mess up? |
| 14:34.07 | mjordan | heh |
| 14:34.08 | mjordan | so! |
| 14:34.49 | mjordan | we were doing some testing and found that loading a config file with catgories with the same name was eating up the CPU. Asterisk was getting into a loop |
| 14:35.00 | mjordan | turns out, it was occurring due to r425384 |
| 14:35.22 | mjordan | calling ast_variable_retrieve while browsing categories is unsafe, if you have multiple categories with the same name |
| 14:35.40 | gtjoseph | rats. |
| 14:35.57 | gtjoseph | got a specific test case? |
| 14:36.09 | file | it changes the internal state to the first occurence, causing a future ast_category_browse to go back to the beginning to find the category |
| 14:36.43 | gtjoseph | ok, looking now. |
| 14:37.29 | file | what's more fun is when you allocate objects based on that and it goes into a loop |
| 14:37.35 | file | RAM? who needs THAT |
| 14:37.39 | file | my RAM is in the cloud |
| 14:37.44 | leifmadsen | and plentiful |
| 14:37.59 | *** join/#asterisk-dev sgriepentrog (~sgriepent@nat/digium/x-txefgnbpvrdaflef) |
| 14:38.48 | gtjoseph | is the situation bad enough that the patch needs to be backed out or can I have an hour to fix and test? |
| 14:39.12 | file | an hour is fine |
| 14:39.18 | gtjoseph | i'm on it then. |
| 14:39.18 | file | or two, or three |
| 14:39.40 | mjordan | really, mostly just looking to have things in good shape by cob tomorrow so we feel good walking into AstriCon |
| 14:40.27 | gtjoseph | gotcha |
| 14:43.42 | file | corn on the cob. |
| 14:45.44 | gtjoseph | file: do you have a code snippet that causes the issue? |
| 14:45.58 | file | not at my hands immediately but I can craft something potentially |
| 14:46.00 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 14:46.11 | gtjoseph | let me see if I ca reproduce first. |
| 14:47.59 | file | http://pastebin.com/UkWSh2Vp |
| 14:48.03 | file | that's pretty much what it does |
| 14:48.15 | file | but you have to have two cats with the same name |
| 14:48.47 | gtjoseph | gotcha |
| 14:49.26 | Marquis42 | Who would name two cats the same, that's crazy pants. ;) |
| 14:50.08 | gtjoseph | Well, it's not like they come when called |
| 14:50.23 | Marquis42 | lol, true that |
| 14:51.06 | *** join/#asterisk-dev sgriepentrog (~sgriepent@nat/digium/x-rrnksthrrildqzjb) |
| 15:04.12 | *** join/#asterisk-dev sruffell (~sruffell@asterisk/the-kernel-guy/sruffell) |
| 15:04.12 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 15:31.32 | *** join/#asterisk-dev sgriepentrog (~sgriepent@nat/digium/x-qqgoplldmynzywxk) |
| 15:38.16 | gtjoseph | file: https://reviewboard.asterisk.org/r/4089/ has the fix. |
| 15:38.34 | file | cool |
| 15:38.56 | file | easy to reproduce? |
| 15:39.12 | gtjoseph | yep and I could see exactly why |
| 15:39.29 | file | ah you did that |
| 15:39.31 | gtjoseph | it's that weird logic in ast_category_browse that tries to "improve performance" |
| 15:40.06 | gtjoseph | I should have chucked it |
| 15:42.21 | file | well, it also allows multiple categories with same name to work |
| 15:42.38 | file | or did. |
| 15:43.32 | gtjoseph | yeah but there was no need to maintain state |
| 15:44.01 | malcolmd | change in 4089 resolves the situation for me |
| 15:44.30 | gtjoseph | and still gives the correct results I hope. |
| 15:46.50 | malcolmd | oh wait..hah, now that you mention it... |
| 15:50.56 | gtjoseph | it doesn't. |
| 15:51.04 | malcolmd | right |
| 15:53.39 | file | I have too much going on. |
| 15:54.03 | file | okay! |
| 15:54.04 | file | yeah |
| 16:05.55 | gtjoseph | mjordan: try it now. |
| 16:11.48 | file | malcolmd, he meant you |
| 16:12.05 | gtjoseph | oops, yeah |
| 16:12.25 | file | looks like it should work |
| 16:12.43 | gtjoseph | i found another case though |
| 16:12.50 | gtjoseph | ast_variable_browse |
| 16:13.19 | file | what about it? |
| 16:13.49 | gtjoseph | it's going to have the same problem where only the first matching category will ever be used. |
| 16:14.01 | ovoshlook | Hello I try to continiue my dialplan afetr exit from stasis through channels/continiue to context:mycontext, extension:myext and priority:mypriority ({"context":"queue.dial","extension":"5000","priority":"4"}). When I POST this I have 204 message but nothing at the cli. At dialplan this context presetns with this priority. |
| 16:14.37 | malcolmd | yup, tried it, better behavior this time. asterisk is still running and i can see the things i should be able to see |
| 16:14.45 | file | gtjoseph, it does not - exactly |
| 16:14.56 | malcolmd | (i think) |
| 16:16.01 | file | if the category passed in is not the last thing browsed then yes |
| 16:16.07 | gtjoseph | file: yeah it will IF you have a category name that you didn't get from ast_category_browse |
| 16:16.24 | gtjoseph | because ast_category_get doesn't use a poinbter compare |
| 16:16.30 | gtjoseph | only a string compare |
| 16:17.57 | file | did the old code have the same problem? |
| 16:18.18 | file | I'm more concerned with ensuring behavior matches previous at this point in time |
| 16:18.30 | gtjoseph | no because it was using the pointer compare first and only did a string compare if it couldn't find a pointer match |
| 16:19.00 | gtjoseph | me too. that's why I didn't rip out the string pointer compare stuff |
| 16:31.17 | mjordan | opticron: does r425691 need to go into trunk? |
| 16:32.09 | opticron | yes, I apparently didn't commit after I merged... |
| 16:32.29 | opticron | done |
| 16:50.16 | *** join/#asterisk-dev philipp64 (~chatzilla@66-232-81-197.ctcweb.net) |
| 17:18.40 | gtjoseph | file: take a look at https://reviewboard.asterisk.org/r/4089/ now. should be good. |
| 17:18.53 | gtjoseph | malcolmd: ^^ |
| 17:20.05 | file | ah |
| 17:29.09 | *** join/#asterisk-dev philipp64 (~chatzilla@66-232-81-197.ctcweb.net) |
| 17:32.42 | file | I'll take a looksee at https://reviewboard.asterisk.org/r/4073/ if anyone wants a PJSIP review |
| 17:35.08 | *** join/#asterisk-dev philipp64 (~chatzilla@66-232-81-197.ctcweb.net) |
| 17:35.08 | Marquis42 | file: Not that any kind of compatibility with chan_sip is a real concern, but any reason why you didn't use the usereqphone like it does? |
| 17:35.36 | Marquis42 | "user_eq_phone is easier to read" is an acceptable answer. ;) |
| 17:35.55 | mjordan | heh |
| 17:35.55 | file | it's easier to read and it also has meaning when it produces CLI and AMI output, it makes it UserEqPhone |
| 17:36.02 | file | unless I made that up |
| 17:36.05 | mjordan | no, you didn't |
| 17:36.10 | mjordan | we conver snake_case to CamelCase |
| 17:36.13 | mjordan | er, convert |
| 17:36.13 | file | yeah. |
| 17:36.50 | Marquis42 | +1, totally reason enough for me. :) |
| 17:39.28 | file | mjordan, merci |
| 17:43.21 | *** join/#asterisk-dev sruffell (~sruffell@asterisk/the-kernel-guy/sruffell) |
| 17:43.22 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 17:49.27 | leifmadsen | huh... snake_case |
| 17:49.30 | leifmadsen | notes this |
| 17:49.48 | file | leifmadsen, careful! it bites |
| 17:50.56 | leifmadsen | oic |
| 17:52.22 | file | Marquis42, merci |
| 17:52.29 | Marquis42 | np |
| 17:56.17 | leifmadsen | mjordan: ping! re: 1.8 |
| 17:56.58 | leifmadsen | mjordan: I note that 1.8 is security-only mode as of AstriDevCon next week. I can't remember, but does that basically mean no more bug fixes allowed, or no more bug fix work done by Digium? |
| 17:57.48 | leifmadsen | I assume that is widespread since Digium would then only release security versions, not bug fix versions, regardless of who was doing the work |
| 17:58.06 | *** join/#asterisk-dev philipp64 (~chatzilla@66-232-81-197.ctcweb.net) |
| 17:58.40 | leifmadsen | so with that being said, does that mean the last 1.8 bug fix should be available sometime during AstriDevCon? (I only ask because I need to get our call servers "updated" and would prefer to just use the last 1.8 version since we'll need to run through regression testing anyways) |
| 18:11.34 | mjordan | no more bug fixes for anyone, since it's a project determination |
| 18:11.51 | mjordan | I mean, you can merge them in - we don't disable commits - but not much point since we won't release it |
| 18:12.01 | mjordan | (hell, you can probably still merge to 1.4) |
| 18:12.18 | mjordan | I'm planning on cutting the first rc of the last version in the next few days |
| 18:12.27 | mjordan | with a release sometime around AstriCon/immediately following |
| 18:14.22 | leifmadsen | mjordan: thanks! |
| 18:14.37 | Marquis42 | Yeah I think he was wondering if we had enough time to clean up and submit a couple of bug fixes to res_jabber |
| 18:14.46 | *** join/#asterisk-dev otherwiseguy (~otherwise@24.27.227.0) |
| 18:15.13 | leifmadsen | Marquis42: actually, I was just more curious about how long I needed to distract management with shiny things for the next rebase :) |
| 18:15.14 | *** join/#asterisk-dev philipp64 (~chatzilla@66-232-81-197.ctcweb.net) |
| 18:15.26 | file | SQUIRREL! |
| 18:15.26 | Marquis42 | lolz |
| 18:15.39 | leifmadsen | Marquis42: you're also not wrong |
| 18:15.46 | Marquis42 | heh |
| 18:27.39 | malcolmd | gtjoseph: thanks :) |
| 18:48.09 | *** join/#asterisk-dev terryw (~otherwise@24.27.227.0) |
| 19:09.19 | coreyfarrell | once 1.8 is in security fix only mode what would security releases be based on? IE /branches/1.8 or /tags/1.8.latest? if it is based on the branch it seems we would want to prevent bug fixes from being merged |
| 19:10.47 | *** join/#asterisk-dev philipp64 (~chatzilla@66-232-81-197.ctcweb.net) |
| 19:12.24 | mjordan | coreyfarrell: security releases are always based on the last released tag |
| 19:12.35 | mjordan | the branch is basically a NoOp at that point. |
| 19:13.17 | mjordan | rarely - although it does happen - we've also made a regression release after the last release IF there was some egregious thing we put into the last release. In that case, as well, it is based on the last made tag |
| 19:13.29 | coreyfarrell | cool I assumed that but had to raise the question just in case I was wrong |
| 19:13.59 | mjordan | no worries :-) |
| 19:14.20 | gtjoseph | can someone take a look at res_stasis_recording in 12 lines 307+ and confirm a suspicion for me? |
| 19:14.25 | mjordan | probably not documented on wiki.asterisk.org. Most of this goofiness is handled by the mkrelease script in repotools |
| 19:14.31 | mjordan | looks |
| 19:15.01 | mjordan | gtjoseph: DTMF termination? |
| 19:15.07 | gtjoseph | yep |
| 19:15.46 | gtjoseph | the NONE and ANY defines are negative but terminate_on is signed. Throws a compile error on Arm. |
| 19:18.03 | mjordan | hm. |
| 19:18.12 | *** join/#asterisk-dev leifmadsen (~free-mads@asterisk/documenteur-extraordinaire/blitzrage) |
| 19:18.12 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 19:18.50 | coreyfarrell | mjordan: while you are here, I was looking into installing refcounter.py to /var/lib/asterisk/scripts/, one is that MENUSELECT_CFLAGS is not available from the main makefile. I'm thinking of creating a very basic contrib/Makefile, adding contrib to OTHER_SUBDIRS in the main makefile. not sure how else to only install refcounter.py if REF_DEBUG is enabled. thoughts on this approach? |
| 19:19.33 | mjordan | coreyfarrell: that sounds reasonable |
| 19:20.55 | mjordan | gtjoseph: that seems a little odd, since terminate_on is a signed char, which can accept negative values. What's the error kicked back from the ARM compiler? |
| 19:21.01 | coreyfarrell | awesome.. I'll try to get a review up for that later tonight. |
| 19:21.29 | gtjoseph | yeah, i think it's a compiler error but i wanted to make sure... error: case label value is less than minimum value for type |
| 19:21.48 | mjordan | hm. Sounds like it thinks that's an unsigned char |
| 19:21.57 | gtjoseph | yeah |
| 19:22.33 | gtjoseph | i'll goof with it |
| 19:24.43 | mjordan | I'm guessing we went with negative numbers there to never conflict with an ASCII code |
| 19:24.53 | mjordan | generally, given what are valid DTMF codes, you could probably pick something else |
| 19:26.06 | gtjoseph | well, it works on x86 so I'm not sure I want to change it. Let me figure out why the switch thinbks it's unsigned. |
| 19:38.05 | gtjoseph | mjordan: default char on arm is unsigned. Duh. I can fix it locally with -fsigned-char |
| 19:47.01 | mjordan | cool |
| 19:47.05 | mjordan | (and that is weird) |
| 19:47.19 | mjordan | Actually, that's super weird. |
| 19:47.37 | gtjoseph | supposedly has to do with the history of the arm platform |
| 19:48.02 | gtjoseph | I'll add the flag in configure.ac so it's set by default for arm. |
| 19:57.50 | file | grrrrr |
| 19:58.13 | gtjoseph | is that aimed at me? |
| 19:59.34 | file | no |
| 19:59.48 | file | should it be? |
| 20:00.03 | gtjoseph | eh, no. Nothing to see here. |
| 20:00.37 | Marquis42 | Sounds like a guilty conscience. ;) |
| 20:01.05 | gtjoseph | well, i HAVE caused enough problems today... |
| 20:01.09 | Marquis42 | heh |
| 20:01.39 | leifmadsen | gtjoseph: it's only 4pm |
| 20:01.58 | leifmadsen | plenty of time to fuck more shit up :) |
| 20:01.58 | gtjoseph | yeah but it's only 2pm my time. A lot can still happen. |
| 20:02.05 | coreyfarrell | gtjoseph: maybe you can use AC_C_CHAR_UNSIGNED instead (or a similar test). this would catch other platforms that might be like ARM if they exist. |
| 20:02.10 | leifmadsen | especially if you started drinking at noon |
| 20:02.43 | gtjoseph | coreyfarrell: yep, that's where I was headed. |
| 20:04.41 | coreyfarrell | gtjoseph: looks like AC_C_CHAR_UNSIGNED doesn't do exactly what we'd want.. probably better to just grab the AC_LANG_BOOL_COMPILE_TRY code it uses and use that to decide if we need to append CFLAGS. |
| 20:05.22 | gtjoseph | looking |
| 20:05.42 | coreyfarrell | on the other hand, it might be worth finding out if -fsigned-char is harmful on "normal" platforms. if it's a no-op on platforms that are signed by default then maybe you could just add it to the default CFLAGS |
| 20:06.06 | gtjoseph | true |
| 20:06.24 | gtjoseph | on intel it doesn't do anything |
| 20:06.30 | gtjoseph | i can test sparc as well |
| 20:19.29 | *** join/#asterisk-dev philipp64 (~chatzilla@66-232-81-197.ctcweb.net) |
| 20:23.20 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-175-118.mci.googlefiber.net) |
| 21:22.05 | *** part/#asterisk-dev mjordan (~mjordan@nat/digium/x-bmpbsgoaumctxvaz) |
| 21:30.13 | *** join/#asterisk-dev philipp64 (~chatzilla@66-232-81-197.ctcweb.net) |
| 22:26.30 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-175-118.mci.googlefiber.net) |
| 23:17.55 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@64.235.216.2) |
| 23:21.58 | *** part/#asterisk-dev kharwell (~kharwell@nat/digium/x-sammlyqesmyxlbcw) |