lx3jl / xlxd Goto Github PK

Are there any resources to help me learn more about XLX 2.0? I want to learn the supported protocols and about the protocol conversion server (like DMR to analog or D-STAR).

Is there a yahoo or google group? What about a Facebook or Google+ page? A wiki, anything?

Currently we have three reflectors and it's getting to complex to manage them. I'm hoping we can do everything with one XLXD 2.0 reflector.

Thanks in advance,
Michael AF6FB

It would be nice with a parrot / echo function for testing

//TSC

There are several sockets that xlxd can't open. The firewall has been set properly and no other programs are using those sockets.

Thank you
73
Ted

Hi
I'm sysop of xlx090, it is installed in a vps in a computer farm with Debian OS. I noticed a strange issue, some repeaters without any apparently problem, during a qso is disconnected from xlx and reconnected after few seconds, I ask if there is some parameters to set to avoid this issue.
Thank you
73 de IW9GRL

Hi,

One of our repeaters is connecting through 3G to the internet. The thrpughput is quite good (approx 1.5Mbps) but ping is ranging between 160 to 360ms
Any stream coming with a ping above 200ms is closed. This seems to be set through the STREAM_TIMEOUT constant.
I cranked this to 2 seconds, and this seems to help with the issue.
Still needs more testing.

BTW G4KLX is using 2 seconds in ircddbgateway for reflector timeouts.

73
Geoffrey F4FXL / KC3FRA

The more I test the YSF feature, the more I desire an option in main.h to just connect YSF client connections to a defined module in main.h. I cannot believe this is not possible.

Existing version of XLX Interlink has requirement to pass D-STAR and DMR traffic at the same time.
Interlink can be forced to do D-STAR to DMR transcoding by downgrading version number.
Please add capability to force DMR to D-STAR conversion.

It would be nice to use hostname instead of IP in case the AMBEd is hosted on a external host with dynamic IP

//SM0TSC - Johan

Sometimes a client opens a stream on the xlxd server with sid 0. This triggers a networkloop with connected ircddbgateway clients. Every client re-transmits this packet and the xlxd server gets "bombarded" with streams. However I think this is actually a bug in ircddbgateway it affects xlxd as well. xlxd crashes within a second after this "loop" starts.

Log entries for clarification:

Apr 10 12:49:32 colo xlxd: Opening stream on module B for client PD4RS E with sid 0
Apr 10 12:49:32 colo xlxd: Detected stream loop on module B for client PA3CQJ C with sid 0
Apr 10 12:49:32 colo xlxd: Detected stream loop on module B for client PA3CQJ C with sid 0
Apr 10 12:49:32 colo xlxd: Detected stream loop on module B for client PA3CQJ C with sid 0
[ .. skip ahead 1000 similar lines ]
Apr 10 12:49:32 colo xlxd: Detected stream loop on module B for client PD4FMJ C with sid 0
Apr 10 12:49:32 colo xlxd: Detected stream loop on module B for client PD4FMJ C with sid 0
Apr 10 12:49:32 colo xlxd: Closing stream of module B

and this is the moment it crashes. The process ends and everyone gets disconnected (about 40 clients). This crash doesn't occur always. Maybe a minimum amount of clients needs to be online to be this destructive.

Is it possible to build in a "safeguard" that drops/discards streams with sid 0? sid 0 seems to be a faulty stream id anyway so I suggest ignoring and discarding those streams. It will help a lot.

Regards,

Remco / PA4TW

I have seen this
Jun 25 22:31:41 localhost kernel: xlxd[13765]: segfault at 8 ip 0000000000428270 sp 00007f1ed432ac40 error 4 in xlxd[400000+68000]
from time to time on XLX010 any ideas?

//SM0TSC

Hi all,
i just updated my XLX146 server .After some testing that i've done locally, at my home network, i am trying now to change the reporting to true; and add the new dashboard address(instead the private IP i had for testing) but it does not seem to work?!No change to the Reflector list, any ideas please? The reflectors works ok (xlx146.ddns.net).

73 de 5B4AMD

I have seen this issue twice.
I have 2x DVsick33 setup as a transcoder in my home, while the actual reflector sits in a data center.
Sometimes the transcode to accept the transcoding, entering some kind of fail state. On the transcoder side, the streams show up, but there is an error on the XLX side.

Under these conditions, only the D-Star headers are sent by the XLX, while the actual DV stream is missing (at least in Dextra and Dplus). This manifests itself as nothing to be heard, with a short appearance of the callsign data at the end of transmission.
After transcoder software restart, everything returns to normal.

I haven't had the time to pursue this issue deeper, but shouldn't the reflector function normally and just ignore transcoding failures, while keeping up regular client to client D-star reflector traffic?

Log of the moment the failure appears:
Jun 24 01:18:18 vps62907 xlxd: Opening stream on module C for client DONGLE B with sid 58321
Jun 24 01:18:27 vps62907 xlxd: Closing stream of module C
Jun 24 01:18:27 vps62907 xlxd: ambed stats (ms) : 0.0/45.3/58.8
Jun 24 01:18:32 vps62907 xlxd: ambed openstream ok
Jun 24 01:18:32 vps62907 xlxd: Opening stream on module C for client DONGLE B with sid 57167
Jun 24 01:19:01 vps62907 xlxd: Closing stream of module C
Jun 24 01:19:01 vps62907 xlxd: ambed stats (ms) : 0.1/46.1/58.3
Jun 24 01:19:01 vps62907 xlxd: ambed 10 of 1452 packets timed out
Jun 24 01:19:13 vps62907 xlxd: ambed openstream ok
Jun 24 01:19:13 vps62907 xlxd: Opening stream on module C for client DONGLE B with sid 22251
Jun 24 01:20:10 vps62907 xlxd: Closing stream of module C
Jun 24 01:20:10 vps62907 xlxd: ambed stats (ms) : 0.1/45.5/62.4
Jun 24 01:20:10 vps62907 xlxd: ambed 5713 of 2808 packets timed out
Jun 24 01:20:17 vps62907 xlxd: ambed openstream ok
Jun 24 01:20:17 vps62907 xlxd: Opening stream on module C for client DONGLE B with sid 50100
Jun 24 01:21:31 vps62907 xlxd: Closing stream of module C
Jun 24 01:21:31 vps62907 xlxd: ambed 14271 of 3648 packets timed out
Jun 24 01:21:36 vps62907 xlxd: ambed openstream ok
Jun 24 01:21:36 vps62907 xlxd: Opening stream on module C for client DONGLE B with sid 36732
Jun 24 01:22:09 vps62907 xlxd: Closing stream of module C
Jun 24 01:22:09 vps62907 xlxd: ambed 6294 of 1626 packets timed out

Luc,

Have a user - W1NCR - registred at DMR and DSTAR and cannot access the reflector because his record is not on your dmrid.dat.

Could you please check whats happen.

thanks

Many Pi-Star users have no idea what they are doing, or are just lazy. They set up Id=XXXYYYY01 in their mmdvm.ini, which works for brandmeister, but often forget to remove the suffix for their xlx connections.

I have a quick hack in my git, which is of course wrong and incomplete, but at least xlxd doesn't fill up my syslog now.

On outgoing DCS streams, the last frame is not encoded as a last frame.
Cause: incorrect conditions in cdcsprotocol.cpp, method HandleQueue:

            if ( packet->IsDvFrame() )
            {
                EncodeDvPacket(m_DvHeadersCache[iModId], (const CDvFramePacket &)*packet, m_iSeqCounters[iModId]++, &buffer);
            }
            else if ( packet->IsLastPacket() )
            {
                EncodeDvLastPacket(m_DvHeadersCache[iModId], (const CDvFramePacket &)*packet, m_iSeqCounters[iModId]++, &buffer);
            }

Last packet is a DvFrame, so the else condition will never be reached.

Corrected code:

            if ( packet->IsDvFrame() )
            {
                if ( packet->IsLastPacket() )
                {
                    EncodeDvLastPacket(m_DvHeadersCache[iModId], (const CDvFramePacket &)*packet, m_iSeqCounters[iModId]++, &buffer);
                }
                else
                {
                    EncodeDvPacket(m_DvHeadersCache[iModId], (const CDvFramePacket &)*packet, m_iSeqCounters[iModId]++, &buffer);
                }
            }

Patch file attached.

cdcsprotocol.diff.zip

The stream is closed right after a close frame is received and this unsets the associated source station of the stream. Right after the close the close frame is sent to all stations which are subscribed to the stream, including the source station, so it receives it's own frame.

Hi,

First of all, thanks to the XLXD authors and contributors.

I have been studying the MMDVM protocol and observed that the XLXD implementation does not match the specification. If possible could it be updated to be more inline with the spec?

MSTPONG

Should be MSTPONG + Repeater ID. Currently the repeater ID is not the valid repeater ID. It seems to be random bytes for me.

RTPACK

Should be RPTACK + Repeater ID. Currently the repeater ID is missing.

MSTNAK

Should be MSTNAK + Repeater ID. Currently the repeater ID is missing.

MSTCL

Should be MSTCL + Repeater ID. Currently the repeater ID is missing.

Thanks

I would like to install xlxd under /opt/xlxd but no luck.

Would appreciate if the system path are configurable via startup parameter, or introduce some config file would be great.

I am trying to have a live display on our dashboard of whether the current talker is using DMR or DStar, and I can't find a reliable method. I can't use the callsign of their repeater or gateway, as some users have the same callsign and band for 2 different connections - DMR and DStar. Can the syslog entries please have the mode somewhere in the entry? Or is there another way to tell?

Jul 1 20:58:15 xlx xlxd: Opening stream on module A for client ZL2EZ B with sid 3150

As we for several reasons have changed the software on DCS010 to XLX we need to be able to have alias DCS010 in parallel to the XLX name (XLX698/DCS698/XRF698) as XLX010 already was taken.. As of now user conecting to the right IP but wrong DCSid (of reflector eg DCS010) is flapping (Rapidly connect / disconnect)

//SM0TSC

Hi,
It maybe will nice to add to XLX protocol SVXReflector

https://github.com/sm0svx/svxlink/tree/master/src/svxlink/reflector

to realize interlink between XLX and SVXReflector

it is part of svxlink (echolink) software. If we will be connected XLX with svxreflector maybe we can
have gateway FM Analog svxlink network with Digital Voice Networks ???

Resolved by Jean Luc

This issue applies to dashboard2 only.

I believe that the "Modules" table in users.php is intended to list user's nodes belong to the module under its column.

For example: Assuming there are two Modules in use, Node 1 & 2 is connected to Module A, while Node 3 on Module B

Module A   Module B
-------------------------
Node 1       Node 3
Node 2

However, the current code will display all the users, in the first column like this.

Module A   Module B
-------------------------
Node 1       
Node 2
Node 3

I managed to fixed the issue with the following change, please take look to see if it is appropriate.
Thanks!

Regards,
Yngwie Chou

=================================================

$ diff -u users.php users.php.fixed
--- users.php   2016-07-15 20:18:25.000000000 +0800
+++ users.php.fixed 2016-07-16 00:03:48.000000000 +0800
@@ -70,26 +70,39 @@

 echo '</tr>';

+$ModuleUsers = array();
 for ($i=0;$i<count($Modules);$i++) {
-
-    $Users = $Reflector->GetCallSignsInModules($Modules[$i]);
-
-   $UserCheckedArray = array();
-
-   for ($j=0;$j<count($Users);$j++) {
-      if (in_array($Users[$j], $UserCheckedArray)) {
-         $CurrentPositions = array_keys($UserCheckedArray,$Users[$j]);
-         $Displayname = $Users[$j].'-'.$Reflector->GetSuffixOfRepeater($Users[$j], $Modules[$i], max($CurrentPositions)+1);
-      }
-      else {
-         $Displayname = $Users[$j].'-'.$Reflector->GetSuffixOfRepeater($Users[$j], $Modules[$i]);
-      }
-      echo '
-            <tr>
-               <td><a href="http://www.aprs.fi/'.$Displayname.'" class="pl" target="_blank">'.$Displayname.'</a> </td>
-            </tr>';
-      $UserCheckedArray[] = $Users[$j];
-   }
+   $Users = $Reflector->GetCallSignsInModules($Modules[$i]);
+   $ModuleUsers[$i] = $Users;
+}
+
+// Find the max node count per module
+$MaxNodeRows = 0;
+foreach ($ModuleUsers as $mUsers) {
+   if ($MaxNodeRows < count ($mUsers)) {
+       $MaxNodeRows = count ($mUsers);
+   }
+}
+
+for ($k=0;$k<$MaxNodeRows;$k++) {
+   echo '<tr>';
+   for ($i=0;$i<count($Modules);$i++) {
+       $UserCheckedArray = array();
+       if ($ModuleUsers[$i][$k]) {
+           if (in_array($ModuleUsers[$i][$k], $UserCheckedArray)) {
+               $CurrentPositions = array_keys($UserCheckedArray,$ModuleUsers[$i][$k]);
+               $Displayname = $ModuleUsers[$i][$k].'-'.$Reflector->GetSuffixOfRepeater($ModuleUsers[$i][$k], $Modules[$i], max($CurrentPositions)+1);
+           }
+           else {
+               $Displayname = $ModuleUsers[$i][$k].'-'.$Reflector->GetSuffixOfRepeater($ModuleUsers[$i][$k], $Modules[$i]);
+           }
+           echo '<td><a href="http://www.aprs.fi/'. $Displayname .'" class="pl" target="_blank">' . $Displayname . '</a></td>';
+           $UserCheckedArray[] = $ModuleUsers[$i][$k];
+       } else {
+           echo '<td></td>';
+       }
+   }
+   echo '</tr>';
 }

 ?>
================================================================

Is possible to bind the server to a specific interface instead of an ip?

This could allow clients from different subnet to connect xlxd.

I have regular client like peers from internet ad other (some repeaters) reachable only via LAN (wifi ham network).
And the gateway is in another subnet than the repeaters. So I have to choose to start xlxd on the ip of the wifi network or the one with public access. If instead I could launch xlxd binded to the interface it could be reachable indistinctly.

308/5000
Hello,
I have done the installation with the readme
nothing appears in the process list and the xlxd.xml is not created.

root @ s265269: ~ # service xlxd start
Starting xlxd:
root @ s265269: ~ #

the server is a V-Server with Debian 7.11 64bit

I have no idea what's wrong

greetings stefan

Hi Luc, I am xlx332 and xlx023 's sysop.
I have compiled and installed the new version of xlx, but the "start service" command does not start. Xlx023 is installed on a datacenter and is currently down.
73 de IZ7JWR Enrico.

TL;DR
It would be useful for xlxd to allow outbound DPLus connections for connection to REFxxx reflectors.

The traditional DPLus reflectors were designed to be a closed world. Some of the dxrfd implementations seem to allow outbound linking via DPlus and I use this to link to an REF reflector on 2 channels concurrently. It works fine. But xlxd is the future. It would be helpful to me to enhance xlxd with the capability to initiate outbound linking to a DPlus based REF reflector. Multiple channel links should be supported. As an example, on my existing dxrfd I link two channels to a DPlus reflector.

This feature could also be useful in overcoming the difficulty posed in initiating two concurrent outbound DExtra links between the same pair of reflectors on different modules. DPlus could be used instead.

The power of xlxd is in its ability to bridge between multiple protocols. Extending this bridge to legacy platforms as well would simplify my reflector infrastructure.

Hi,
after the update/upgrade of the new xlxd and dashboard(done from scratch) we are trying to connect to the xlx145 and we are getting every 30 seconds approx a disconnect packet. Why is this? The connection between the 2 reflectors prior to the update was ok with no issues at all. All the settings were tripled checked from our side xlx146. At the moment the service is up and running and we deleted the interlink entry to avoid connection-disconnection every 30 seconds waiting to find a solution. The admin from 145 states that it is possibly their fault and has something to do with the upgrade. Please see the log below.

Feb 20 18:12:30 xlx146 xlxd: XLX ack packet for modules AC from XLX145 at 178.59.21.32
Feb 20 18:12:30 xlx146 xlxd: New peer XLX145 at 178.59.21.32 added with protocol 4
Feb 20 18:12:30 xlx146 xlxd: New client XLX145 at 178.59.21.32 added with protocol 4 on module A
Feb 20 18:12:30 xlx146 xlxd: New client XLX145 at 178.59.21.32 added with protocol 4 on module C
Feb 20 18:12:47 xlx146 xlxd: XLX disconnect packet from XLX145 at 178.59.21.32
Feb 20 18:12:52 xlx146 xlxd: XLX (1.4.2) connect packet for modules AC from XLX145 at 178.59.21.32
Feb 20 18:13:01 xlx146 xlxd: XLX peer XLX145 keepalive timeout
Feb 20 18:13:01 xlx146 xlxd: Client XLX145 at 178.59.21.32 removed with protocol 4 on module A
Feb 20 18:13:01 xlx146 xlxd: Client XLX145 at 178.59.21.32 removed with protocol 4 on module C
Feb 20 18:13:01 xlx146 xlxd: Peer XLX145 at 178.59.21.32 removed

Any ideas of what it is causing 145 to send a disconnect packet every approx 30 seconds? Just to add that now the reflector is connected to xlx888 which as i 've noticed is periodically disconnecting (not every 30 seconds) as well as other reflectors. Please visit xlx888 to see what i mean.

Thanks in advance for your time.

73 de 5B4AMD-George
owner/host of xlx146

Hi Luc,

i think that it is important for the future to have the possibility to choose the Module which is been transcoded.
Every stream right now is send to the AMBE-Server and so i get stuck with only 4 Channels available.

Bye
Timm

Problem and solution in one:

Problem:
With more and more services which are connected to XLX by transcoders, a repeater callsign may connect several times parallel.
XLX assigns a constant module ID to DMR conects, by default "B".
This creates issues if you are connecting with a real DSTAR repeater which uses the same callsign.
"B" was always reserved for DSTAR repeaters on UHF.
I am running several repeaters which use DSTAR on 70cm (B), 23cm (A), 10m (E), DMR and YSF.

Different DMR devices (or YSF2DMR etc.) with the same callsign all get the same Module ID.

Solution:
I created a little patch which creates the Module-ID based on the DMR-ID.
It starts with "H" and by that reserves A to G for real DSTAR modules.
The ID is calculated from the DMR-ID, so it is always the same for the same DMR-ID.

I placed a patch to a fork at https://github.com/dl5di/xlxd
The changes are in the file https://github.com/dl5di/xlxd/blob/master/src/cdmrmmdvmprotocol.cpp

73 de Hans DL5DI

While testing the new build of the LX3JL USB-3006 device, under ambed 1.3.1 observed two conditions:

1. If I change the product description string from USB-3006 to XC-3006 and modify the cftdidevicedescr.cpp to recognize that string, on startup it gets an error:

Opening USB-3006_A:(serial number) device
FTDI function FT_OpenEx error 2

2. If additionally 2 ThumbDVs are plugged in along with the USB-3006, it sees both the XC-3006 and ThumbDV devices, but only reads the USB-3006 and never completes initializing the server. (Is this a feature or bug?) It works fine with the 2 ThumbDVs or USB-3006 independently.

The good news is the sample production unit is performing well with the software. We are still doing some hardware testing in the lab, but if no issues are found, the production run should ship right after Hamvention (watch the NW Digital Radio blog at http://nwdigitalradio.com )

Hi everyone,

I have an xlx with ambed software on a second machine, when the connection between xlxd and ambed starts to have a very variable latency the streaming ambe xlx side crashes and no longer resumes, the only way is to restart xlxd.

By increasing the TRANSCODER_AMBEPACKET_TIMEOUT and STREAM_TIMEOUT the thing gets better but the fact is that when it crashes it does not recover.

Adding some prints in the file ctranscoder.cpp, forcing the printing of ambed stats and packet timeouts, noticed that the ping goes to -1000.0 and changes the order of the printouts, but the keepalive communication remains sent, the operation of which remains unchanged. If anyone wants to take a look I attach log files.

Could it be some problem of managing any counter or lack of management of a state?

And some time beating myself but have not come to terms with it yet.

IZ1MLT Luca

log xlxd.txt

$PageOptions['PageRefreshActive'] = true; // Activate automatic refresh

Is set to true

//Johan - SM0TSC

As soon as I add a DV-Stick 30 Stick to the USB Port of the AMBE Server the AMBE3003 stops workling.
Both Ambes are recognized but no channels available.
As soon as I reject the DV-Stick 30 all is ok again.

... not the module the user was last heard on.

Hi Luc.

I have just created a xlx server for Denmark. XLX238. But its still not listed in the reflector list. Can you please add it ?

thanks

73 - Lasse oz6nor.

Is the software capable of configuring multi ambe transcoders with different ip addresses?

module A <-> ext ip 12.x.x.2
module B <-> ext ip 15.x.x.3

if so, how do I set this up?

the password was hardcoded in the log php file and I could access most of the xlx server and get the content of /var/log/messages.

The column "IP" on the peers page of the dashboard contains a hyper link to the IP address of the reflector in case the IP addresses (or parts of it) are not hidden. That is not pointing to the dashboard in every case.
For XLX518 for example this points to a completely different web page (the one of our amateur radio club) which does not directly relate to the XLX reflector or its dashboard.
Instead it would be nice to be able to reach the dashboard of the connected peers. Therefore I made a little patch. See #10
The code from my branch can currently be viewed at: http://xlx518.n18.de/index.php?show=peers

As stated in the title, tx.gif is missing in the /img folder of dashboard2 and needs to be copied from dasboard.

It would be nice if we could link into Brandmeister talkgroups on DMR native (not via DCS or other D-STAR protocol).. This could be done (probably) by using the info here

https://wiki.brandmeister.network/index.php/Simple_External_Application

//SM0TSC - JOhan

Hi,

Setup a new reflector today xlx530.ddns.net
I can see the dashboard is live and reflector showing in the hostfile lists.
But nothing happens when trying to connect to it via any hotspot or repeater.
Using a static IP address, and port forwarding is all set up correctly.
Any advice here would be appreciated?

Cheers
Daniel ZL4DE

There is an issue with the fact that if there are queued packets in an module at the end of a transmission, the transmitting client will get back a short sequence of its own Dv packets.
This is not a major issue, but it has some side effects:
In DCS, since each packet carries complete header info, it will match the users own info which may show up on the transceiver screen.
In Dextra mode, since the last few frames don't have header infosm it will put a blak message on screen. This is most disturbing e.g. on the UP4DAR, where it will erase the screen at the end of an incoming transmission.

Hi,

I've noticed that dmrid.dat was never re-loaded, it seems that it is required to restart xlxd in order to accept connections via mmdvm protocol from clients that were not defined in dmrid.dat previously

http://paste.debian.net/1000930/ is a quick and dirty hack that works for me. it would be nice if reloading of dmrid.dat can be implemented properly.

A user from the United States with a callsign beginning with KC4 connected and is showing to be from Antartica. According to this Wikipedia page, calls with the suffix beginning with "A" or "U" are assigned to Antartica. https://en.wikipedia.org/wiki/Amateur_radio_call_signs

At this moment XLX interlink uses D-STAR headers only to pass traffic between reflector. If user has more than one ID a reflector system on the other side may use another ID than used on original call.

Here's scenario:

For DExtra Protocol, clients are usually using callsign XRFooo for reflector connection.
However, iif a G4KLX client is using XLXooo callsign instead of XRFooo in DExtra_Hosts.txt and connecting to XLX reflector, then client will be dropped due to lack of keepalive.

Here's the log when a client is using XLXooo callsign in DExtra_Hosts.txt for reflector connection.

/opt/XLX# tail -f /var/log/messages
Jul 17 10:50:21 localhost xlxd: DExtra client BV5OO   C keepalive timeout
Jul 17 10:50:21 localhost xlxd: Client BV5OO   C at 10.100.1.100 removed
Jul 17 10:51:10 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 10:51:10 localhost xlxd: New client BV5OO   C at 10.100.1.100 added with protocol 1
Jul 17 10:51:10 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 10:55:04 localhost xlxd: DExtra client BV5OO   C keepalive timeout
Jul 17 10:55:04 localhost xlxd: Client BV5OO   C at 10.100.1.100 removed
Jul 17 10:55:57 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 10:55:57 localhost xlxd: New client BV5OO   C at 10.100.1.100 added with protocol 1
Jul 17 10:55:57 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 10:56:58 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 10:56:58 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 10:57:58 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 10:57:58 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 10:58:58 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 10:58:58 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 10:59:58 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 10:59:58 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 11:00:58 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 11:00:58 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 11:01:58 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0
Jul 17 11:01:58 localhost xlxd: DExtra connect packet for module C from BV5OO   C at 10.100.1.100 rev 0

To me, it seems the xlxd is using callsign to tell if the incoming connection is a peer or a client.
Not sure if this will be a security concern.

73,
Yngwie
BX2AFC

At random, our xlxd 2.2.2 sometimes stops passing audio, and requires a service restart to solve. It might go 15 days without this behaviour, or it might be 5 hours. No clues in the syslog, everything looks normal.

Ubuntu 18.04 LTS
Intel Atom d525mw Dual Core
4GB RAM (3.5GB free)

I have git cloned, and re-compiled from scratch, and still having the same problem.

Existing implementations of CPacketStream and CPacketQueue don't care about probabilistic reordering of UDP packets. Normally it should care about sequence numbers to process audio frames on the right order.