2026.05.10/05.25/05.30 - Package Updates - Beta Channel

We continue to release updates to the ASL3 project. We start off sharing these updates with our developers, then to the “beta” package channel for those who want to test fixes to their problems or get the “latest and greatest” as quickly as possible. When we believe that the changes are OK for all we push them out to the production channel.

Before opting to install the "beta" (or "devel") packages we ask that you carefully read Beta Testing about the caveats and potential pitfalls of beta packages.

The following ASL3 packages are now available from the "beta" channel. The following are highlights of the changes:

asl3 3.18-2

  • asl-node-auth-check: node changed to s in lookup PR #183
  • asl-node-auth-check: Handle missing nodes and add ability to follow #tryinclude PR #184
  • asl-node-auth-check: Add a few more null checks PR #185
  • CodeRabbit Generated Unit Tests: Add unit tests PR #186
  • Add CI workflow for running tests with Python PR #187
  • Change CI branch from 'main' to 'develop' PR #188
  • Add shell script test platform PR #189
  • Add test_play_arn() PR #191
  • Refactor test into directories PR #192
  • asl-say: Add tests PR #193
  • 100% test coverage for asl-play-arn PR #194
  • asl-node-auth-check: Tighten up key failures PR #195
  • Add permissions to workflow PR #196

asl3-asterisk 22.9.0+asl3-3.9.0-1

  • asterisk

    • Update from Asterisk 22.8.2 to Asterisk 22.9.0
  • asl3-asterisk

    • Fix spelling; coupled with app_rpt change to rpt_telemetry.c PR #81
    • Restore tranceive.ulaw (when needed) PR #83
    • asl3-asterisk-config.postinst: handle \r\n (in .conf files) PR #84
  • app_rpt

    • app_rpt: Add IS_ECHOLINK_NODE macro PR #986
    • app_rpt: ast_copy_string() assures null termination PR #987
    • app_rpt: a bit of logging and a bunch of cleanup PR #988
    • tests: Update from DAHDI to Local channel PR #989
    • app_rpt: Address various memory leaks PR #990
    • app_rpt: Allow nodelog to dump logged messages to the file system on app unload PR #991
    • app_rpt: Add all missing curly braces to if statements PR #992
    • app_rpt: Add curley's step 2 PR #993
    • xmpr.c: Add {} to if statements PR #994
    • correct spelling; corresponding change is coupled in asl3-asterisk/rpt-sounds PR #995
    • app_rpt: address memory leak on load_rpt_vars() PR #997
    • app_rpt: Handle freeing of mdc memory PR #998
    • app_rpt: Handle telemetry datastore leaks on failure PR #999
    • pocsag.c: Limit buffer process size PR #1002
    • app_rpt.c: Address potential buffer overrun in rpt_do_sendtext() PR #1004
    • extensions.conf: radio-secure context use NODENUM PR #1006
    • app_rpt.c: Use ARRAY_LEN() in pocsag PR #1007
    • app_rpt.c: Add RPT_CONFIG_VAR_INT_MIN_FLOOR macro PR #1008
    • app_rpt.c: Add RPT_CONFIG_EXPLODE_STRING macro PR #1009
    • app_rpt.c: Common format for AST_LIST_TRAVERSE PR #1011
    • app_rpt.c: Missing {} from PR #993 PR #1012
    • app_rpt.c: xpmr missing curly PR #1013
    • app_rpt.c: Full format from clang-format configuration PR #1014
    • app_rpt: code reformatting (mostly white space) PR #1019
    • app_rpt: no need to pass "size - 1" when API includes NUL term PR #1020
    • Coderabbit Updates PR #1021
    • app_rpt: Collection of CodeRabbit "finds" PR #1022
    • app_rpt: Update README.md PR #1023
    • chan_tlb: Address truncated node number PR #1028
    • Remove dead code and convert comments to C style PR #1030
    • chan_usrp: process text frames before voice frames PR #1033
    • chan_echolink: start el_directory thread after last possible failure path PR #1035
    • rpt_serial: replace for() loops with usleep() for parallel port timing PR #1037
    • rpt_gps: failover to general_def_position if either coordinate is missing PR #1038
    • app_rpt: Correctly handle ast_dsp_process() frames PR #1040
    • chan_echolink: Remove intermediate queue and general refactor for multi thread PR #1042
    • chan_echolink: restore ability to make outgoing calls PR #1046

These packages will be available for the next 1-2 weeks. If we do not hear of any issues or problems with the updates we will promote them to the production/main channel.

Again, please see Beta Testing for information on how to use package software and the caveats and warnings about beta test software.

Sincerely,
AllStarLink

The 2026.05.10 beta release appears to have fixed the above referenced issue, where Echolink users hear connected ASL nodes with choppy audio.

However, the same problem now exists in the opposite direction. Echolink users now sound choppy when being heard on the ASL nodes.

This has been confirmed on both a VPS hosted node and a Raspberry Pi 5 hosted node.

Eric
K2CB

Do you have anything special in your environment such as a VPN tunnel or an unusually small MTU size on the Interface? Is there a threshhold number of direct-connect clients (Echolink + IAX) that seem to trigger the issue?

I am using a VPN tunnel between sites. Tailscale. Been using for a year or so. No issues until the 04.09.2026 upate. The MTU settings are all default. I am not a networking expert, so if there is something to be concerned about MTU settings, I am willing to try adjusting them, with some guidance. There is no correlation as to how may Echolink + IAX clients are connected, as far as I can tell. It occurs with only one Echolink user connected to an ASL/Echolink node, with this node and a repeater node both connected to the same hub node, so I would assume that is considered to be one IAX connection and one Echolink connected at the affected node.

Here is how I tested things -

I have a test hub node 672017 which is ASL only. I then have node 672019 which has Echolink node 468347 "married" to it via the echolink.conf settings. I used 672017 as a "radioless hub", connecting 672019 to 672017. I then connected a third ASL node with a simpleusb repeater connected to it, node 672007.

If I talk on a radio, which is received by the repeater/node 672007, and connect the Echolink program on my notebook pc to Echolink node 468347, and listen to the radio transmission, the audio received on Echolink sounds great (this used to be the problem prior to the 2026.05.10 update).

If I talk on the Echolink program on the notebook pc, the listeners on the repeater (node 672007) report that my audio is very choppy (this direction used to sound fine prior to the update).

The nodes 672017 and 672019 are running on a VPS. The repeater node 672007 is running on a Pi5.

I have a duplicate setup running for a friend's system, with the Echolink/ASL node 668179, repeater node 668172, and "radioless" hub node 668170. It experiences the same exact issue.

This occurs even with only one Echolink user connected to the ASL/Echolink node.

If anyone cares to hear it, they are welcome to connect to hub node 672000 and Echolink node 468347.

FWIW, in order to placate the many Echolink users on this system, even though I am not a fan, I spun up a Pi4 running HamVoIP with Echolink, ASL node 616530 / Echolink node 21005, which is basically mimicking the affected ASL3 node 672019 / Echolink 468347 setup. The HamVoIP setup (616530) is then also connected to the hub node 672000. Someone can be talking on any of the other ASL3 nodes connected to hub node 672000, and the Echolink users report great audio, with no issues, when connected to the setup running on the HamVoIP installation. The same users can then immediately switch to the ASL3 setup, and the audio issues appear.

Eric
K2CB

Is it possible to test without Tailscale running? There's something specific here in the chan_echolink driver that's interacting only with certain types of systems. Many people didn't have this problem before and still don't. I run a large hub with many Echolink and IAX Direct clients who don't experience this. So there's something niche going on we need to troubleshoot and nail down.

*** UPDATE 2026.05.25 ***

An EchoLink issue was reported with the "beta" packages first shared on 2026.05.10. Changes to correct this issue have been included in a set of updated packages. The changes include:

asl3-asterisk 22.9.0+asl3-3.9.2-1

  • app_rpt

    • chan_echolink: use read to process audio PR #1048
    • chan_echolink: missing locks on continue PR #1050
    • chan_echolink: Don't use translated frame (restore original behavior) PR #1053

These packages will be available for the next 2-3 weeks. If we do not hear of any issues or problems with the updates we will promote them to the production/main channel.

Again, please see Beta Testing for information on how to use package software and the caveats and warnings about beta test software.

Sincerely,
AllStarLink

*** UPDATE 2026.05.30 ***

One more set of updates to correct EchoLink audio issues have been included in the "beta" packages. The changes include:

asl3-asterisk 22.9.0+asl3-3.9.3-1

  • app_rpt

    • chan_echolink.c: Use 20ms timer to buffer variable inbound gsm frames PR #1056

These packages will be available for 1-2 weeks. If we do not hear of any issues or problems with the updates we will promote them to the production/main channel.

Please see Beta Testing for information on how to use package software and the caveats and warnings about beta test software.

Sincerely,
AllStarLink