| 02:16.41 | *** join/#asterisk-dev CoderForLife (~Miranda@cpe-174-101-155-51.cinci.res.rr.com) |
| 02:43.54 | *** join/#asterisk-dev tzafrir_laptop (~tzafrir@bzq-218-155-148.cablep.bezeqint.net) |
| 02:44.00 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 02:48.01 | *** join/#asterisk-dev elguero (~miguel323@ns1.nashuacs.com) |
| 03:32.04 | *** join/#asterisk-dev kaber (~kaber@HSI-KBW-109-192-188-061.hsi6.kabel-badenwuerttemberg.de) |
| 03:35.10 | kaber | I'm looking for some guidance .. I've written a DECT channel driver which needs to store data for dynamically registered portable phone, like the allocated regextension, authentication key etc. The ast_db interface seems quite inflexible for this since its text-based, can't store complete records without encoding them into a single string etc. What is the recommended method of doing this in a channel driver? |
| 03:38.14 | Corydon76-dig | kaber: sounds like a good application of a realtime driver |
| 03:39.01 | Corydon76-dig | check out include/asterisk/config.h, in particular the ast_store_realtime and ast_destroy_realtime |
| 03:39.01 | kaber | my understanding is that to use realtime drivers the data has to be looked up on every use, is that correct? |
| 03:39.11 | kaber | thx I'll have a look |
| 03:39.40 | Corydon76-dig | For configuration, the more you lookup from a db, the more realtime the data is |
| 03:40.01 | Corydon76-dig | but for certain reasons, a lot of drivers cache information from the database |
| 03:40.09 | kaber | basically I want to store the data on registration and reload it on startup, but use it directly from mmory |
| 03:40.33 | Corydon76-dig | You can use the interface for that, I suppose... |
| 03:40.44 | Corydon76-dig | and update the values, too? |
| 03:40.44 | kaber | yeah, it just felt a bit like abuse :) |
| 03:40.47 | kaber | yes |
| 03:41.12 | kaber | the main reason for loading it into memory is that the portables should be available for cli-commands |
| 03:41.33 | kaber | so I don't only want to react on incoming or outgoing call |
| 03:42.18 | Corydon76-dig | I don't see a problem with doing it that way |
| 03:42.26 | kaber | ok thanks, I'll give it a try |
| 03:43.02 | Corydon76-dig | One question... if the backend is not there, will your driver fail horribly? |
| 03:43.18 | Corydon76-dig | Or just have to load from the config file each time? |
| 03:43.21 | kaber | not horribly, but the phones would have to reregister |
| 03:43.42 | Corydon76-dig | We have a table in chan_sip called sipregs for that purpose |
| 03:43.43 | kaber | the config file includes only a few static settings, most data originates from pairing |
| 03:44.13 | Corydon76-dig | sippeers is for loading configuration; sipregs is for storage of information received back from sip peers |
| 03:44.26 | kaber | that sounds similar to what I need |
| 03:44.47 | kaber | can the realtime backends store complex records? |
| 03:44.48 | Corydon76-dig | Yep, sounds like it |
| 03:45.12 | kaber | or do I need to encode them in a single string similar to the ast_db interface? |
| 03:45.27 | Corydon76-dig | You have multiple fields, but a single row |
| 03:45.51 | Corydon76-dig | If you want more complex than that, you need to build out the relationship yourself |
| 03:45.55 | kaber | I see, just found the code |
| 03:46.29 | Corydon76-dig | ast_update_realtime for keying off a single field; ast_update2_realtime for keying off multiple fields |
| 03:46.53 | kaber | that looks like it will work |
| 03:47.33 | *** join/#asterisk-dev tuxd00d (~tuxd00d@69-92-218-70.cpe.cableone.net) |
| 03:47.44 | kaber | one more question while I'm at it - my drivers needs root privileges to communicate with the kernel part of the DECT stack - which doesn't work when the driver is loaded late since astrisk has already dropped privileges |
| 03:47.53 | kaber | is there a way to regain root temporarily? |
| 03:47.58 | kaber | I couldn't find anything .. |
| 03:48.08 | Corydon76-dig | One sec |
| 03:50.26 | Corydon76-dig | The way to do this is with capabilities |
| 03:50.57 | Corydon76-dig | Or just set up the device to be owned by the Asterisk user with udev permissions |
| 03:51.09 | kaber | its using netlink, so there is no device node |
| 03:51.14 | kaber | but yeah, CAP_NET_ADMIN would work |
| 03:51.28 | Corydon76-dig | No need, then. CAP_NET_ADMIN is already held |
| 03:51.43 | Corydon76-dig | We use it for setting TOS bits on IP packets |
| 03:51.45 | kaber | mhh that's odd, I get a permission denied when loading the module |
| 03:51.54 | kaber | guess I need to debug it some more |
| 03:52.57 | kaber | ah ok that was my mistake, the socket layer requires cAP_NET_RAW for certain things |
| 03:53.02 | kaber | I guess I'll change that |
| 03:53.05 | Corydon76-dig | Ah |
| 03:53.17 | Corydon76-dig | It's in main/asterisk.c |
| 03:53.22 | kaber | thanks |
| 03:53.48 | Corydon76-dig | Make sure that you give it all 3, eip, so it can retain privs on restart |
| 03:54.19 | kaber | i think I'll change the kernel instead to also accept CAP_NET_ADMIN for opening these sockets |
| 03:56.05 | kaber | ok thanks I guess I'll get to work on these things :) |
| 03:56.20 | kaber | will probably submit the entire thing in a month or two |
| 03:59.00 | Corydon76-dig | Cool |
| 05:51.46 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 05:51.46 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 06:30.03 | *** join/#asterisk-dev hehol (~hehol@buero-gw.dortmund.loca.net) |
| 07:00.46 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 07:00.46 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 07:10.01 | *** join/#asterisk-dev tamiel (~tamiel@213.30.183.226) |
| 08:49.46 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 08:49.46 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 10:06.26 | *** join/#asterisk-dev ghenry (~ghenry@pdpc/supporter/monthlybyte/ghenry) |
| 10:06.29 | ghenry | Hi, |
| 10:06.37 | ghenry | I've been out of the loop a while |
| 10:06.39 | ghenry | svn: OPTIONS of 'https://origsvn.digium.com/svn/asterisk/trunk': SSL handshake failed: SSL alert received: Certificate is expired (https://origsvn.digium.com) |
| 10:09.22 | *** join/#asterisk-dev bklang (~bklang@tesla.alkaloid.net) |
| 10:10.40 | kpfleming | you need to contact jpeeler to get your certificate renewed |
| 10:34.13 | ghenry | OK, thanks |
| 10:38.34 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 10:38.34 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 11:16.29 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 11:16.29 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 11:20.39 | ghenry | What's jpeeler email? |
| 11:20.45 | ghenry | jpeeler@digium.com ? |
| 11:22.59 | kpfleming | yes |
| 11:27.23 | ghenry | thanks |
| 12:13.29 | *** join/#asterisk-dev hugorebelo (~hugo@200-171-132-124.completo.com.br) |
| 12:19.31 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 12:19.31 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 12:20.03 | *** join/#asterisk-dev guax (~guax@unaffiliated/guaxinim) |
| 12:21.22 | guax | i dont know if this is worth here on -dev but ill try anyways, i collected a log from AMI on asterisk 1.4.30 and this is what came up: http://pastebin.com/xZ7YfASD its a atxfer and i cant tell why that link/unlink happens so many times |
| 12:33.04 | pabelanger | guax: #asterisk |
| 12:36.57 | *** join/#asterisk-dev hugorebelo (~hugo@200-171-132-124.completo.com.br) |
| 13:09.20 | *** join/#asterisk-dev anonymouz666 (~anonymouz@187-28-37-118.poolip.RJO.embratel.net.br) |
| 13:34.16 | *** join/#asterisk-dev jpeeler (~jpeeler@asterisk/digium-software-dev/jpeeler) |
| 13:34.16 | *** mode/#asterisk-dev [+o jpeeler] by ChanServ |
| 13:55.02 | *** join/#asterisk-dev leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 13:55.02 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 14:12.17 | pabelanger | The '!' cli command, asterisk just spawns a shell correct? |
| 14:12.44 | *** join/#asterisk-dev bklang (~bklang@tesla.alkaloid.net) |
| 14:13.49 | pabelanger | and then I assume if you restart asterisk, your in for a world of hurt? |
| 14:30.46 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 14:30.46 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:31.03 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 14:31.17 | *** join/#asterisk-dev elguero (~miguel323@12.187.84.162) |
| 14:32.45 | *** join/#asterisk-dev Deeewayne (~dwayne@75-150-14-49-Atlanta.hfc.comcastbusiness.net) |
| 14:32.45 | *** mode/#asterisk-dev [+o Deeewayne] by ChanServ |
| 14:34.21 | *** join/#asterisk-dev dms (~dms@173-10-22-217-BusName-utah.ut.hfc.comcastbusiness.net) |
| 14:51.59 | Qwell | seanbright: you cheeky bastard! |
| 14:52.06 | Qwell | M17380 |
| 14:52.08 | MuffinMan | [assigned] [Asterisk] CDR/cdr_tds 0017380: [patch] long waiting (over 2 min) for cdr_tds running "SELECT 1 FROM cdr" when asterisk starting with aournd 3mil cdr in table reported by kkwong https://issues.asterisk.org/view.php?id=17380 |
| 14:52.11 | Qwell | I *love* that patch. |
| 14:56.10 | putnopvut | Holy dongs that is weird |
| 14:56.22 | putnopvut | But effective! :) |
| 14:59.01 | Qwell | having run into that same problem with other stuff, I know exactly why it works too |
| 15:01.26 | *** join/#asterisk-dev The_Boy_Wonder (~vossel@asterisk/batman-developer/dvossel) |
| 15:01.26 | *** mode/#asterisk-dev [+o The_Boy_Wonder] by ChanServ |
| 15:07.46 | seanbright | heh |
| 15:13.08 | seanbright | don't have a ton of time today if one of you gents want to commit that |
| 15:17.07 | bklang | I'm working on larger columns right now |
| 15:17.22 | bklang | whoops, wrong -dev |
| 15:45.53 | *** join/#asterisk-dev sruffell (~sruffell@asterisk/the-kernel-guy/sruffell) |
| 15:45.53 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 16:01.17 | *** join/#asterisk-dev tuxd00d (~tuxd00d@69-92-218-70.cpe.cableone.net) |
| 16:04.39 | *** join/#asterisk-dev Victor_Yure (~aaa@unaffiliated/victoryure/x-837844) |
| 16:06.51 | *** join/#asterisk-dev marv (~timr@router.asteriasgi.com) |
| 16:27.38 | *** join/#asterisk-dev paravoid (~paravoid@zebra.cslab.ece.ntua.gr) |
| 16:42.07 | *** join/#asterisk-dev sgimeno (~chatzilla@21.Red-88-27-235.staticIP.rima-tde.net) |
| 17:23.21 | *** join/#asterisk-dev elguero (~miguel323@12.187.84.162) |
| 17:25.10 | *** join/#asterisk-dev tzafrir (~tzafrir@bzq-218-155-147.cablep.bezeqint.net) |
| 17:25.10 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 17:35.45 | *** join/#asterisk-dev mpe (~mpe@85.24.87.252.generic-hostname.arrownet.dk) |
| 17:54.59 | *** join/#asterisk-dev mnicholson (~mnicholso@nat/digium/x-syfckojfybydrbnz) |
| 18:01.06 | *** join/#asterisk-dev jeffg (~jeffg@pdpc/supporter/professional/jeffg) |
| 18:28.27 | *** join/#asterisk-dev mrgabu (~gbdurante@187.38.158.209) |
| 18:29.02 | guax | !faltam |
| 18:29.12 | guax | shit, wrong channel :P sorry |
| 18:42.09 | *** join/#asterisk-dev philipp64 (~chatzilla@66.232.79.143) |
| 18:44.33 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 18:51.04 | dms | pabelanger: Hey Paul, I went through and made all the changes per your feedback to the app_swift module and uploaded a new diff to the review board and mantis. |
| 18:56.18 | philipp64 | russellb: let me know if you ever need to schedule some time to look at the bamboo/cross-build issues. |
| 19:00.21 | Corydon76-dig | philipp64: I'm looking at cross build issues now |
| 19:00.51 | Corydon76-dig | philipp64: but I'm stuck on your build environment, because netsnmp is currently stuck |
| 19:01.04 | Corydon76-dig | and Asterisk won't build without termcap support |
| 19:01.32 | *** join/#asterisk-dev ctooley (~ctooley@c-98-193-232-154.hsd1.tn.comcast.net) |
| 19:01.32 | *** mode/#asterisk-dev [+o ctooley] by ChanServ |
| 19:01.53 | Corydon76-dig | Please to be fixing, kthx |
| 19:06.58 | philipp64 | yeah, that's a known issue... and indeed there's a bug opened for that as well. |
| 19:07.19 | philipp64 | M16991 |
| 19:07.21 | MuffinMan | [acknowledged] [Asterisk] Core/BuildSystem 0016991: Can't build asterisk using just --with-netsnmp... seems to always want to use CONFIG_NETSNMP reported by pprindeville https://issues.asterisk.org/view.php?id=16991 |
| 19:07.38 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 19:07.38 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 19:08.24 | philipp64 | have a look at ~philipp/kernel/reast or just modify the .config file to have BR2_PACKAGE_NETSNMP= |
| 19:12.55 | Corydon76-dig | philipp64: No, I mean that your build system won't build Asterisk until it gets past building netsnmp |
| 19:15.00 | philipp64 | yup, that's another known issue... netsnmp doesn't build against a current version of lmsensors. We've filed a bug. do this: |
| 19:15.18 | philipp64 | build everything until it fails... then restart the build as: |
| 19:15.44 | philipp64 | make netsnmp-dirclean netsnmp BR2_PACKAGE_LMSENSORS= ; make all BR2_PACKAGE_NETSNMP= |
| 19:16.10 | philipp64 | which builds netsnmp without linking to lmsensors, and builds everything else (including asterisk) without linking to netsnmp. |
| 19:17.41 | Corydon76-dig | Where in your build system is the Asterisk build stuff defined? |
| 19:18.22 | *** join/#asterisk-dev bencer (~bencer@heal.cauterized.net) |
| 19:20.09 | bencer | hi all, what has changed on app_queue what is not getting the member interface from res_config_ldap ? i get app_queue.c:1428 create_queue_member: No location at interface '' but some debug on res_config_ldap says that the interface is ok: res_config_ldap.c:995 realtime_multi_ldap: ldapdebug: SIP/foo |
| 19:31.10 | philipp64|laptop | Corydon76-dig: sorry, where is which stuff defined? |
| 19:31.32 | philipp64|laptop | it's a distributed in a couple of places, so you'll have to be more precise. |
| 19:34.25 | philipp64|laptop | some of the info comes from .config, and some of it comes from package/asterisk/asterisk.mk |
| 19:42.02 | philipp64|laptop | Corydon76-dig; why is BUILD_CC needed if we already have HOST_CC? |
| 19:43.08 | philipp64|laptop | in addition to having CC_FOR_BUILD ... |
| 19:43.22 | Corydon76-dig | philipp64|laptop: because in this environment, HOST_CC is the wrong one |
| 19:44.02 | Corydon76-dig | HOST_CC would only be correct if you were building a cross-compiler |
| 19:44.18 | philipp64|laptop | HOST_CC should be the "native" compiler, not the one for the machine on which the binaries will eventually run. |
| 19:44.20 | *** join/#asterisk-dev mpe (~mpe@85.24.87.252.generic-hostname.arrownet.dk) |
| 19:44.30 | Corydon76-dig | philipp64|laptop: that's incorrect |
| 19:44.55 | philipp64|laptop | well, that's how it is in the 138 other projects we build. |
| 19:45.09 | Corydon76-dig | I refer you to the autoconf documentation: http://www.gnu.org/software/hello/manual/autoconf/Specifying-Target-Triplets.html#Specifying-Target-Triplets |
| 19:45.32 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 19:45.32 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 19:46.35 | Corydon76-dig | The host is the same as the build ONLY when you're either NOT cross-compiling or when you're compiling a cross-compiler |
| 19:47.23 | Corydon76-dig | CC_FOR_BUILD and BUILD_CC are the same thing |
| 19:47.31 | Corydon76-dig | or same concept, anyway |
| 19:49.04 | philipp64|laptop | ok, so rhino, oslec, openssl, etc. all do things different. they use HOSTCC |
| 19:50.02 | philipp64|laptop | everyone has a different way to do it. |
| 19:50.02 | Corydon76-dig | I'll go with official gnu documentation in this case |
| 19:50.21 | philipp64|laptop | you'll be in the minority, but that's fine. |
| 19:51.03 | Corydon76-dig | Given that we're using gnu autoconf, I'm going with a concept that makes it less confusing for me. |
| 19:52.15 | philipp64|laptop | ok, so what needs to be passed to configure? |
| 19:52.28 | Corydon76-dig | --host=triplet |
| 19:53.09 | Corydon76-dig | Set your PATH beforehand so that <triplet>-gcc is in PATH |
| 19:54.06 | Corydon76-dig | You can additionally specify --build=triplet, if you want, but it's generally already correct (based upon output of config.guess) |
| 19:55.05 | philipp64|laptop | well, currently configure looks at CC_FOR_BUILD,HOST_CC,CC ... so will those go away from the argument list to configure? |
| 19:55.38 | Corydon76-dig | philipp64|laptop: Curl's configure help agrees that the CC is build, not host: |
| 19:55.40 | Corydon76-dig | <PROTECTED> |
| 19:55.40 | Corydon76-dig | <PROTECTED> |
| 19:56.37 | Corydon76-dig | and you're building that, so at least one of your other packages agrees with the gnu documentation |
| 19:56.48 | Corydon76-dig | I think it may be that you're confused, not the packages. |
| 19:57.06 | philipp64|laptop | and yet it works. |
| 19:59.31 | Corydon76-dig | philipp64|laptop: I wonder if that's because curl doesn't need to build anything to run on the build system |
| 20:00.09 | Corydon76-dig | philipp64|laptop: find something else that needs to run something on the build system... oh, like your netsnmp config thingy... |
| 20:00.31 | Corydon76-dig | I wonder if THAT is why it doesn't work... because you built it with the wrong compiler |
| 20:04.11 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 20:04.11 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 20:05.40 | Corydon76-dig | Maybe not... looks like net-snmp-config is a shell script |
| 20:06.04 | Corydon76-dig | But it's probably finding the wrong shell script... because of your PATH |
| 20:43.45 | *** join/#asterisk-dev ccesario (~ccesario@189-19-6-236.dsl.telesp.net.br) |
| 20:53.44 | *** part/#asterisk-dev soul_of_guax (~guax@unaffiliated/guaxinim) |
| 21:18.31 | philipp64|laptop | you mean it was finding the build systems |
| 21:18.44 | philipp64|laptop | 's version of netsnmp-config? correct. |
| 21:19.25 | philipp64|laptop | indeed, this is a recurring problem in a lot of build systems. what would be ideal is to pass the information directly to "configure" that would normally be gleaned from netsnmp-config. |
| 21:19.43 | philipp64|laptop | that is, making any sort of pkgconfig script a "last resort" if the information hasn |
| 21:20.00 | philipp64|laptop | hasn't been explicitly given (I hate this macbook keyboard). |
| 21:20.17 | Corydon76-dig | philipp64|laptop: yes, but your build system created its own config script that has the correct values |
| 21:20.36 | Corydon76-dig | All it needs is for the PATH to allow it to be found first |
| 22:10.40 | philipp64|laptop | that gets really tricky... |
| 22:11.00 | philipp64|laptop | because you have no way of differentiating scripts from binaries. |
| 22:11.36 | philipp64|laptop | and whatever you pass in the PATH to configure also affects how configure runs, i.e. which version of "ls" it uses, etc. |
| 22:12.41 | philipp64|laptop | plus the version of netsnmp-config in ./build_i586/root/usr/bin/netsnmp-config assumes that it's being run chroot'd in ./build_i586/root/ which of course it's not. |
| 22:57.36 | *** join/#asterisk-dev mpe (~mpe@0x4dd624b2.adsl.cybercity.dk) |
| 23:06.45 | *** join/#asterisk-dev tuxd00d (~tuxd00d@69-92-218-70.cpe.cableone.net) |
| 23:32.32 | *** join/#asterisk-dev Netgeeks (~chris@gw1.netgeeks.net) |
| 23:41.42 | *** join/#asterisk-dev tzafrir_laptop (~tzafrir@212.179.75.202) |
| 23:41.42 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |