November 2009 - App_rpt-users - lists.keekles.org

First experience with ACID app_rpt & usbradio audio popping problem.
by Steve Gladden 20 Nov '09

20 Nov '09

Short version: *Installed ACID *Have audio popping problem with USB transmit (output) audio. *I'm not surprised didn't expect it to work perfectly first attempt. *Only tried one combinarion of hardware/cpu/motherboard so far.. I need to try other motherboards/CPUs. * Tried 3 different USB devices and ports. * played with moving interrupts disabling motehrboard built-on hardware etc. * CPU runs at about 10% while USB radio is active.. * Loading PC heavily by running gzip -9 seti@home Lots of disk access lots of network activity seem to have zero impact on the popping frequentness. * I'd like to know if anyone else here has had this or similar problems with consistent audio from the USB device (CM-108) and what they may have done to make it work better. Long Version: I gave it a go and went the 'easy' route just to get started.. I ended up trying it out on that 1.1Ghz AMD Duron CPU rack mount server.. I must say I'm extremely excited & please with this software.. But it would be too good to be true if it all just worked first try.. and of course would take all the 'fun' out of getting it to work.. Ok enough small talk (haha) Here's what I have going on: Everything works except I am getting intermittent audio popping in the audio output of the USB sound device. VERY intermittent maybe about 3-4 'pops' every minute on average.. 10 at most sometimes.. and other times several minutes go by without a 'pop'. It seems to mostly occur during the CW ID beacon which I have going every minute.. The pops that are heard only happen when the 'transmitter' is active sending PL or CW ID or PLAYBACK... My test arrangement has "radio key" running to produce continuous PL and occasional audio popping. There is no radio actually hooked up yet. OK will talk about the hardware a bit.. and probably worse case scenario here due to an 'all in one' motherboard.. It's a MSI 6378 'budget' Athlon board.. everything built on. Has USB 1.1 *NOT 2* I don't know/think this matters for a sound device. Here's all of the things I have tried all of which have absolutely no effect on the intermittent popping... 1. Disabling everything not needed on motherboard including ethernet for test.. ethernet + usb share an interrup line and there is no ability to re-route this in the bios so I turned off the ethernet on the motherboard for testing. 2. Carefully monitor CPU during audio pops running 'top' around 20 updates per second (this alone takes more CPU load than asterisk /app_rpt + usb radio. 3. On this motherboard the USB controller and ethernet share a PCI interuupt line and you cannot change it. I've tried disabling the built on ethernet and still get the popping.. I've also flood and large pinged the machine which genreates huge numbers of interrups from the ethernet chip and this has no effect on the popping.. 4. Tried other USB fobs with same chipset. Created various types of heavy system activity to see if it has any impact on the popping heavy disk use hevy network use and heavy vga use (text mode) moving interrupts (changing them) on the motherboard.. getting usb controller onto a different interuupt (ethernet ends up on same). 5. Measured and scoped 5V output on USB to be sure 'popping' sound is not coming from 5V power supply. 6. Am really looking to hear what some others experience has been with this.. what works and what does not? The popping very much sounds like an interrupt issue or some kind of issue where the task of sending and audio stream is completely dropped for something else and then resumed leaving lots of holes heard as pops in the output stream. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.

5 13

The audio file upload feature on allstarlink.org
by Stephen Rodgers 16 Nov '09

16 Nov '09

I coded an addition to allstarlink.org which allows custom connect messages to be uploaded, stored, and eventually re-distributed to all Allstar modes. I need help from a few node operators to upload a .wav file to test the upload procedure. To upload a .wav file, first record your connect message in .wav format sampled at a multiple of 8KHz. (e.g. 48 KHz). The file size should be less than 2 megabytes. You connect message should be short and be something like "WA6ZZZ Repeater, Los Angeles California" Once the .wav file is recorded, you can upload it by logging in to allstarlink.org with your node number and password. There will be a button to to upload the audio file once you are logged in. The button will present a form which you can use to upload the audio file. Use the form and upload your audio file. I would like to know if anyone gets any errors when uploading the .wav file. Pleas make a valid recording for your node so that it can be used when the distribution part is implemented. Steve WA6ZFT

2 1

Re: [App_rpt-users] First experience with ACID app_rpt &usbradio audio popping problem.
by George Csahanin 16 Nov '09

16 Nov '09

Steve- This sounds like it may be the same thing I was reporting, but I still haven't found a way to get a recording of it. http://mantis.qrvc.com/view.php?id=34 Steve Gladden wrote: > Hi David.. it's not a terribly loud popping it's right with the normal > audio > levels > It really sounds like 'holes' in the processing small abrupt dropouts of > the > stream.. > Very random in timing maybe 5-15 hits per minute when I key the radio with > PL.. > > Seems to more frequently happen on/around the CWID but still is happening > very randomly. > > CPU utilization is ~10% while radio keyed producing PL and CWID.. > Have radio continuously keyed (no actual radio) and CDID beacon going > every > minute. > > 5-15 'dropouts' per minute. This seems to describe what I see, but I don't classify it as a pop, but the PL is interrupted and the mobile RX briefly mutes. Take care GeorgeC W2DB George J Csahanin W2DB Cedar Park, TX

1 0

Testing the mailing list
by Stephen Rodgers 16 Nov '09

16 Nov '09

Does it work now?

3 4

Mailing list back up
by Stephen Rodgers 16 Nov '09

16 Nov '09

The IP address for the qrvc.com server was mistakenly classified as dynamic and was policy blocked by Spamhaus (PBL list) at my ISP providing service to my home. This appears to have been resolved. I had not seen any list messages for a few days so I took a look at the logs on the qrvc.com website which told me what was going on. Steve WA6ZFT

1 0

playback cmd?
by John Bettasso 15 Nov '09

15 Nov '09

Hello, Has global playback been implemented? It appears that playback cmd performs the same as localplay cmd by only playing sound files locally. Thanks, John, AJ9ON

1 0

Test. Please disregard
by Stephen Rodgers 15 Nov '09

15 Nov '09

Test. Please disregard

1 0

Unable to connect in to a 5 digit node
by KD8B 13 Nov '09

13 Nov '09

Hello; Looking for a little help again... I set up my new testing node on 27013. I get failed connects inbound but it does look like 27013 is registering correctly... outbound seems to work at least sometimes... If I recall correctly ( from last night.. I was sleepy) 27013 connects to 2235 but not to 2364, neither 2235 or 2364 can call in to 27013... Is there anything we are going to need to fix in extensions.conf or such to provide for the mask change of 4 chars to 5 chars in node name??? or is this some firewall issue again?? Both nodes I am testing with are registered but When connections come in that are not on port 4569... what do they come in on as a range??? maybe I need to forward those through my firewall?? I am using Untangle firewall and port forwarding from anything on the outside interface destined locally for port 4569 to pass through on 4569 to the inner specified host.... I am pretty sure you guys set up 2235 to only use 4569... this may have something to do with it.. In the big picture I Would rather fix the firewall to allow other ports through as needed than lock all to 4569.. but if that is what it takes ..ok :-) Maybe packets arrive on one port and I am forcing them internally throug NAT to 4569 and this is messing something up???? Thoughts???? ON a separate issue ( although it does include the same nodes as involved parties) 27013 is at a separate site from (2235 and 2364) .... both sites are behind a Similarly configured Untangle firewall... I also have 2 machines on 1 computer 2235 and 2364 and I can connect from 2235 to 2364 but not from 2364 to 2235... being both on localhost 127.0.0.1 it would seem to be an internal problem but scanning the 4 conf files I cannot see what.... both have registrations in iax.conf, extensions.conf and in rpt.conf.. both work individually but I am missing this... from extensions.conf [radio-secure] ;exten => ${NODE},1,rpt,${NODE} exten => 2235,1,Rpt,2235 exten => 2364,1,Rpt,2364 **************************** By the way.. are the "r" in rpt here case sensitive ??? I have seen listings both ways************** [radio-gui] exten => 2235,1,Rpt,2235|X exten => 2364,1,Rpt,2364|X from iax.conf register=A2235:xxxxxx@register.allstarlink.org register=A2364:xxxxxx@register.allstarlink.org ; Incoming radio connections [radio] type=user disallow=all allow=g726aal2 allow=gsm codecpriority=host context=radio-secure transfer=no from rpt.conf [nodes] ; Note, if you are using automatic update for allstar link nodes, ; no allstar link nodes should be defined here. Only place a definition ; for your locak nodes, and private (off of allstar link) nodes here. 2235 = radio(a)127.0.0.1/2235,NONE ; This must be changed to your node number 2364 = radio(a)127.0.0.1/2364,NONE ; Memories for remote bases and I have this in line after both node stanza's so it is duplicated twice.. statpost_program=/usr/bin/wget,-q,--timeout=15,--tries=1,--output-document=/dev/null statpost_url=http://stats.allstarlink.org/uhandler.php ; Status updates Any help????? Doug

1 0

Autopatch problems
by Stephen - K1LNX 13 Nov '09

13 Nov '09

I started playing with the autopatch and reverse autopatch functions lastnight and ran into some problems. My setup consists of two seperate boxes, one is running app_rpt/radio only, and I have an Asterisk 1.6/FreePBX install as my home PBX. My ultimate goal is to direct all outgoing autopatch calls to the PBX box and then on to my provider, and have phone control of the radio from it as well. I can't quite put a finger on the best way to accomplish this, but one of the problems I'm having is with the reverse autopatch, for starters. I have this in sip.conf in the app_rpt box: [k1lnx] type=user context=radiocontrol secret=mypassword host=dynamic allow=gsm transfer=no This is in the PBX box in /etc/asterisk/sip_registrations_custom.conf: register => k1lnx:mypassword@192.168.1.1/k1lnx When I issue a SIP reload on both boxes, I can not reg for whatever reason: (FreePBX box) [Nov 12 10:17:02] WARNING[32096]: chan_sip.c:15790 handle_response_register: Got 404 Not found on SIP register to service k1lnx(a)192.168.1.1, giving up (app_rpt box) [Nov 12 10:17:02] NOTICE[23702]: chan_sip.c:15628 handle_request_register: Registration from '<sip:k1lnx@192.168.1.1 <sip%3Ak1lnx(a)192.168.1.1>>' failed for '192.168.1.15' - No matching peer found It's highly possible I have some syntax wrong somewhere and I have not tried pointing a softphone at the app_rpt box yet with these credentials to eliminate the FreePBX box as the issue. My thoughts are if I can get a SIP registration between both boxes I should be able to accomplish autopatch/reverse autopatch very easily, unless there is another easier or preferred method? Thanks, Stephen -- ********************************** Stephen Brown - ARS K1LNX Johnson City, TN EM86 http://www.k1lnx.net google voice: 423-665-9367 **********************************

2 2

Allstar Link going to 5 digit node numbers
by Stephen Rodgers 07 Nov '09

07 Nov '09

In order to conserve the number space, Jim and I have decided to assign 5 digit node numbers starting from the 27XXX series. This will give us up to 3000 additional node numbers in the 2XXXX/2XXX block. No more 4 digit node numbers will be assigned until further notice. Steve WA6ZFT

1 0