- FreeSWITCH-users - FreeSWITCH Mailing Lists

Re: [Freeswitch-users] Running FS on a heavy loaded VMWare host
by Hristo Benev 03 Sep '13

03 Sep '13

Why to not over commit... Actually it is simple and logical... If you commit resources hypervisor will wait for those resources to be available before giving them to the VM. I've already had a host that was sitting idle on CPU, but VMs were laggy. After investigation it appeared to be over commiting of CPU But as I mentioned earlier: Test your setup before putting it in production. >-------- Оригинално писмо -------- >От: EL fs.user(a)fordior.net >Относно: Re: [Freeswitch-users] Running FS on a heavy loaded VMWare host >До: FreeSWITCH Users Help <freeswitch-users(a)lists.freeswitch.org> >Изпратено на: Понеделник, 2013, Септември 2 15:15:33 EEST > > One general tip for optimization - do not assign more resource to a VM than it actually needs. > > Can you tell me why more resources isn't good for optimization? Do you mean > to make efficient use of bare metal? > > -- > EL > > _________________________________________________________________________ > Professional FreeSWITCH Consulting Services: > consulting(a)freeswitch.org > http://www.freeswitchsolutions.com > > FreeSWITCH-powered IP PBX: The CudaTel Communication Server > http://www.cudatel.com > > Official FreeSWITCH Sites > http://www.freeswitch.org > http://wiki.freeswitch.org > http://www.cluecon.com > > FreeSWITCH-users mailing list > FreeSWITCH-users(a)lists.freeswitch.org > http://lists.freeswitch.org/mailman/listinfo/freeswitch-users > UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users > http://www.freeswitch.org ----------------------------------------------------------------- Опитай новия аромат на Armani Code Ultimate мъже за 35лв http://parfiumbg.com/?utm_source=netinfo&utm_medium=abvmail&utm_content=Out…

1 0

Re: [Freeswitch-users] Running FS on a heavy loaded VMWare host
by Hristo Benev 03 Sep '13

03 Sep '13

It is important to define heavily loaded host... What is loaded - network, CPU, RAM, I/O? If you cannot guarantee enough resources for new FS VM probably you will have troubles. Best advice will be to try it yourself. One general tip for optimization - do not assign more resource to a VM than it actually needs. >-------- Оригинално писмо -------- >От: Grant Bagdasarian GB(a)cm.nl >Относно: [Freeswitch-users] Running FS on a heavy loaded VMWare host >До: "FreeSWITCH Users Help (freeswitch-users(a)lists.freeswitch.org)" <freeswitch-users(a)lists.freeswitch.org> >Изпратено на: Понеделник, 2013, Септември 2 09:34:29 EEST > > > > ; > > > > > Hello, > > I’ve been going through some threads about virtualization, some say it works for them, others say it doesn’t. Multiple platforms are used like MS Hyper-V, VM Ware, etc. > We currently run FS on a physical box, which works great. > > We have our own virtual environment, VMWare vSphere 5.0. Our 5.0 environment is running under heavy load and I was wondering would FS (configured properly for correct timing) still function properly and able to handle hundreds of concurrent calls on a heavy loaded VMWare host? The FS virtual will also process RTP in default mode. I was hoping if someone had any experience using this setup and could give me some insights on this matter. The FS virtuals will function only as SBC’s, so no complex dial plans, no call recording, no conferencing. Just bridging the incoming call leg with an outgoing one. > > VMWare has a tech paper about running VoIP on their virtual machine, and they claim it works. (http://www.vmware.com/files/pdf/techpaper/voip-perf-vsphere5.pdf) In the paper they used Mitel vMCD, but I’m not sure if the software has been optimized to run on VMWare so the tests would come out better than running any other telephony software like FreeSwitch or Asterisk. > > Grant > > ----------------------------------------------------------------- Облечи върху тялото си свежият аромат от Burberry Body жени 30лв http://parfiumbg.com/?utm_source=netinfo&utm_medium=abvmail&utm_content=Out…

4 3

can't use a set variable
by David Villasmil 02 Sep '13

02 Sep '13

Hello Guys, I was already told about using the "inline" when setting a variable, which actually did set it, but i can't then use it, for some reason.. This is my dialplan <condition field="destination_number" expression="^.*"> <action inline="true" application="set" data="rand_val=${expr(randomize(&x);ceil(random(0,100,&x)))}"/> <action application="log" data="INFO Random value is ${rand_val}"/> <action inline="true" application="set" data="auth_call=${cond(${rand_val} < 41 ? YES : NO )}"/> <action application="log" data="INFO Result: ${auth_call}"/> <condition field="${auth_call}" expression="^YES$"> <action application="log" data="INFO ASR 40 we got ${rand_val}"/> <anti-action application="log" data="INFO ${auth_call} is not greater than 41"/> </condition> </condition> and the Log: EXECUTE sofia/external/12345(a)1.2.3.4 log(INFO cond(19 < 41 ? YES : NO ) is not greater than 41) 2013-09-02 17:57:27.090531 [INFO] mod_dptools.c:1567 cond(19 < 41 ? YES : NO ) is not greater than 41 EXECUTE sofia/external/12345(a)1.2.3.4 log(INFO Random value is 19) 2013-09-02 17:57:27.090531 [INFO] mod_dptools.c:1567 Random value is 19 EXECUTE sofia/external/12345(a)1.2.3.4 log(INFO Result: cond(19 < 41 ? YES : NO )) 2013-09-02 17:57:27.090531 [INFO] mod_dptools.c:1567 Result: cond(19 < 41 ? YES : NO ) 2013-09-02 17:57:27.090531 [NOTICE] switch_core_state_machine.c:262 sofia/external/12345(a)1.2.3.4 has executed the last dialplan instruction, hanging up. 2013-09-02 17:57:27.090531 [NOTICE] switch_core_state_machine.c:264 Hangup sofia/external/12345(a)1.2.3.4 [CS_EXECUTE] [NORMAL_CLEARING] Apparently the <condition field="${auth_call}" expression="^YES$"> is not working... Can you guys help? Thanks! David

2 1

freeswitch crash on receiving a fax
by Ivan Milivojevic 02 Sep '13

02 Sep '13

Hi all, I have a problem with hylafax and freeswitch for INBOUND faxes.I configured hylafax and one modem (FS0) and everything is fine. root:/var/log# faxstat HylaFAX scheduler on test: Running Modem FS0 (123456): Running and idle root:/var/log# ls -al /dev/FS0 lrwxrwxrwx 1 root root 10 Sep 2 15:41 /dev/FS0 -> /dev/pts/3 Dialplan in FSW is simple, ask lua for status and use available modem.I can try without lua, but i dont think that lua is a problem. <extension name="LUA CALL INBOUND FAX"> <condition field="destination_number" expression="^(123456)$" break="never"> <action application="lua" data="fax_query.lua ${destination_number}" inline="true"/> </condition> <condition field="${status}" expression="aktivan"> <action application="log" data="INFO ACTIVE FAX, STATUS = [${status}]"/> <action application="answer" /> <action application="playback" data="silence_stream://2000"/> <action application="bridge" data="modem/a/a"/> <anti-action application="log" data="INFO NOT ACTIVE FAX"/> <anti-action application="hangup"/> </condition> </extension> The problem is that on receiving fax, FSW crashes.I cannot see EXECUTE bridge... freeswitch@internal> version FreeSWITCH Version 1.2.8+git~20130503T183844Z~65f0d16dee (git 65f0d16 2013-05-03 18:38:44Z) OS is Ubuntu 12.05, I will try this week on debian/centos also. Freeswitch log and syslog are in attach. Can anyone has the idea what can i do? Regards, Ivan

3 2

FS-3918: git commit: f91bdf12aa1854da04775933a40f444e613775b5
by Nuno Reis 02 Sep '13

02 Sep '13

Hi Ken. I've recently upgrade my FS environments to the latest stable 1.2.12. I've found that everytime i entered a conference it resulted in a segmentation fault. When digging on the problem i found that it is related with the commit referenced above. Basically i had to set the conference_moderator_pin channel variable when entering a conference otherwise i have a segmentation fault in FS. Regarding this issue my question is why did you make the mdpin pointer variable a const? I would like to hear about your thoughts on this. Thanks. BR, -- *Nuno Miguel Reis* | *Unified Communication** Systems* M. +351 913907481 | nreis(a)wavecom.pt WAVECOM-Soluções Rádio, S.A. Cacia Park | Rua do Progresso, Lote 15 3800-639 AVEIRO | Portugal T. +351 309 700 225 | F. +351 234 919 191 *GPS<http://maps.google.com/maps/ms?msa=0&msid=202333747613191340808.0004b4b227a…>| www.wavecom.pt** <http://www.wavecom.pt/>* [image: Description: Description: WavecomSignature]<http://www.wavecom.pt/pt/wavecom/premios.php> [image: Publicity] <http://www.wavecom.pt/pt/mail_eventos.php>

3 3

Re: [Freeswitch-users] Running FS on a heavy loaded VMWare host
by Jeff Leung 02 Sep '13

02 Sep '13

Now here comes the fun part -debating what virtualization platform works best for FreeSWITCH. We won't get there just yet ;) In my experience, as long as your hypervisor provides a synthetic timing/clock device, you should be fine as long as the underlying host isn't completely overloaded. From: freeswitch-users-bounces(a)lists.freeswitch.org [mailto:freeswitch-users-bounces@lists.freeswitch.org] On Behalf Of Grant Bagdasarian Sent: Sunday, September 1, 2013 11:34 PM To: FreeSWITCH Users Help (freeswitch-users(a)lists.freeswitch.org) Subject: [Freeswitch-users] Running FS on a heavy loaded VMWare host Hello, I've been going through some threads about virtualization, some say it works for them, others say it doesn't. Multiple platforms are used like MS Hyper-V, VM Ware, etc. We currently run FS on a physical box, which works great. We have our own virtual environment, VMWare vSphere 5.0. Our 5.0 environment is running under heavy load and I was wondering would FS (configured properly for correct timing) still function properly and able to handle hundreds of concurrent calls on a heavy loaded VMWare host? The FS virtual will also process RTP in default mode. I was hoping if someone had any experience using this setup and could give me some insights on this matter. The FS virtuals will function only as SBC's, so no complex dial plans, no call recording, no conferencing. Just bridging the incoming call leg with an outgoing one. VMWare has a tech paper about running VoIP on their virtual machine, and they claim it works. (http://www.vmware.com/files/pdf/techpaper/voip-perf-vsphere5.pdf) In the paper they used Mitel vMCD, but I'm not sure if the software has been optimized to run on VMWare so the tests would come out better than running any other telephony software like FreeSwitch or Asterisk. Grant

3 2

Why doxygen doesn't generate doc for many modules ?
by Jayanth Acharya 01 Sep '13

01 Sep '13

Given the default Doxygen.conf file in docs/ directory, I found that many of the optional modules (even enabled in modules.conf, which of course, I don't expect doxygen to parse/understand), are not included in the documentation. Is it because they are seen as not having any API exposed ? I don't see an explicit list of directories to scan either. What am I missing ? J

1 2

Lua script problem
by Phil Quesinberry 01 Sep '13

01 Sep '13

Hi everyone, Would anyone have any ideas as to why the following statement would not work and end up assigning a null value to scripts_dir? scripts_dir = string.sub(debug.getinfo(1).source,2,string.len(debug.getinfo(1).source)-(st ring.len(argv[0])+1)); I'm pretty sure that this is related to a much larger problem that we're experiencing where MWI has stopped working completely on all extensions. We started experiencing intermittent problems with this after making recent updates to FS and FusionPBX before it stopped working entirely. Marc Crane, the developer of FusionPBX, is under the impression that this is either an FS bug or that something has changed with FS' handling of MWI. I don't know lua so I can't weigh in on this myself. I filed a Jira on this, see the following for more details, including the full script the above excerpt is taken from: http://jira.freeswitch.org/browse/FS-5739 Any input would be greatly appreciated. Phil Quesinberry Q Systems Engineering, Inc. (410) 969-8002 http://www.qsystemsengineering.com <http://www.qsystemsengineering.com/>

3 2

inbound-bypass-media for 1.2 (stable version)
by Nikolay Rogoshchenkov 30 Aug '13

30 Aug '13

Is it possible to use this option <param name="inbound-bypass-media" value="true"/> for 1.2 version? -- Rogoshchenkov Nikolay

3 2

Suggestion for additional syntax for "continue_on_fail"
by Nathan Neulinger 30 Aug '13

30 Aug '13

It seems like there are a bunch of codes that would typically indicate a problem with a gateway, as opposed to a valid return - like no answer or busy. My current list that I'm working with that I'm treating as "upstream failed to handle it for some reason" is: CALL_REJECTED,GATEWAY_DOWN,RECOVERY_ON_TIMER_EXPIRE,NORMAL_TEMPORARY_FAILURE,DESTINATION_OUT_OF_ORDER, NORMAL_UNSPECIFIED,UNALLOCATED_NUMBER,NO_ROUTE_DESTINATION It would be nice to be able to both use negation type syntax, as well as 'collection's. I was thinking something like this: ALL,!USER_BUSY,!NO_ANSWER Example from OpenSSL: SSLCipherSuite ALL:!ADH:!EXPORT56:RC4+RSA:+HIGH:+MEDIUM:+LOW:+SSLv2:+EXP:NULL-MD5:NULL-SHA For now, I'm just using a variable for it defined centrally. -- Nathan ------------------------------------------------------------ Nathan Neulinger nneul(a)mst.edu Missouri S&T Information Technology (573) 612-1412 System Administrator - Architect

2 1