bf124257e8
LINK STALE clearing logic will not fire until 12 hours of stale activity by the current app-level rules.
...
I also updated README.md (line 18) to document the 12-hour threshold.
2026-06-03 17:52:00 -07:00
3d3b48735c
Added a compile-time guard in Transport.cpp:
...
RNS_SUPPRESS_TRANSPORT_STATS
Enabled it for Exercise 205 in platformio.ini:
-D RNS_SUPPRESS_TRANSPORT_STATS=1
2026-06-03 11:56:31 -07:00
23fbc569c6
LINK payloads now look like BOB says Hi to CY iter=0.
...
Sender/recipient use board IDs for easier log aggregation.
Iteration count is still included and increments per peer.
Send cadence is now once per minute at the node’s allocated slot only. Removed the previous slot + 30s second send.
Removed noisy app-internal diagnostics like APP LINK..., APP RX..., and TX LINK SKIP.
Set Reticulum library logging to LOG_WARNING, which suppresses heap/path-store/entries/byte-count style diagnostics while preserving warnings/errors.
Fixed inherited physical slot label Fay to Flo.
Updated the README with the message cadence and examples
2026-06-03 10:00:57 -07:00
2a6fef41c7
What changed from 204:
...
Kept all units in transport mode with reticulum.transport_enabled(true).
Removed physical blocking: SIM_PHY_BLOCK_BOB_CY=0, no per-unit block.
Removed the extra deep transmission/debug flags from platformio.ini.
Kept the 204 announcement schedule/protocol.
Removed intentional Link teardown after message cycles.
Every unit now attempts an outbound Link to every peer that announces.
Added retry health behavior: 3 Link attempts within 3 minutes, then wait for a fresh announce before trying that peer again.
Preserved substantive parseable logs: TX ANNOUNCE, RX ANNOUNCE, TX LINKREQUEST, LINK ACTIVE, TX LINK, RX LINK, retry/failure/reset events.
Updated copied helper scripts to point at 205_sustained_link.
Updated README for the new exercise behavior.
2026-06-03 09:19:33 -07:00
eecd1a9f4c
awk parsers for logs
2026-06-03 09:00:28 -07:00
7d8b3eb705
Adding loader
2026-06-02 18:46:20 -07:00
690c8987b3
Added directory check
2026-06-02 18:25:21 -07:00
237ccf5103
To facilitate deployment among other machines and upload from there directly
2026-06-02 11:42:11 -07:00
db57359abd
Current announce cycle was: once per minute, at 2 + NODE_SLOT_INDEX * 4 seconds. So BOB at :06, CY at :10, DAN at :14, etc.
...
I changed it to:
first announce: immediately after startup once microReticulum is ready
second announce: after ANNOUNCEMENT_2 seconds
later announces: every ANNOUNCEMENT_REPEAT seconds
2026-06-02 11:22:27 -07:00
dc8e6163af
Another delay injection
2026-05-29 18:04:17 -07:00
ce56e876bd
Note tested, added 750ms delay to overcome possible race condition
2026-05-29 17:46:56 -07:00
a52617862f
Before further debugging of LINK bug
2026-05-29 17:24:33 -07:00
4e1d9a75db
For posterity
2026-05-29 12:30:45 -07:00
294a17660d
John draft, I asked Codex to review and it started changing, I intervened and asked to keep Codex review separate, I think it sucessfuly backed out its edits. Committing for posterity, there is a style difference between me and Codex.
2026-05-29 12:30:04 -07:00
197b46b4bb
Exercise is reduced to BOB, CY & DAN. LINK is still broken, but writing history_of_a_LINK.md for a forensic analysis and referencing this version
2026-05-29 10:58:14 -07:00
7f78dfb70e
Change frame to have prefix showing origin. Tested and all units received from the other tree. I should have committed then, but forgot. Here, the blocking has been activated, but not tested.
2026-05-28 21:23:10 -07:00
fd96e56410
All 4 units, BOB, CY, DAN & ED, are LINKing with the other three. Success.
2026-05-28 20:50:16 -07:00
d8e5cc0a8d
Safety
2026-05-28 17:25:31 -07:00
f1670a6123
4 Units are talking, but LINK rarely works. Committing now for safety.
2026-05-28 16:44:41 -07:00
10498b57e1
Additional schema change to reconcile with Perl importer
2026-05-27 11:55:49 -07:00
ff7c6fa0ad
Untested revision of schema as of 11:50 AM
2026-05-27 11:51:30 -07:00
58c62ed07d
Chat GPT revision, small, after 11:00 a.m. today
2026-05-27 11:47:21 -07:00
e555986b87
Improved workflow for migrating logs into database
2026-05-27 11:16:23 -07:00
8a75b96ac5
Working rssi plotly, also saved in Subversion
2026-05-27 11:11:59 -07:00
fe46db2b3c
latest schema for 18 column logs
2026-05-27 11:10:14 -07:00
a8fc2cf9d5
Preserving an earlier version of the schema along with other files
2026-05-27 11:09:24 -07:00
a982223055
Picking up changes I had made before migrating to 18 column
2026-05-26 07:36:04 -07:00
de2262f747
Preserving for posterity before update to 18 column
2026-05-26 07:34:11 -07:00
577caba635
After time changes, but before testing trial #2
2026-05-25 14:56:23 -07:00
aaa765adf8
After trial upon which import_exercise_26_ble_log.pl & schema were created.
2026-05-25 14:10:58 -07:00
b407554210
Prior to GPS code changes
2026-05-25 11:59:09 -07:00
d38c646553
Untested, after completion of first round
2026-05-24 15:40:50 -07:00
49cead9636
the Pi builds work, ready for release
2026-05-22 16:41:10 -07:00
6717b56ffb
Created a debug optional dichotomy, last test run was clean... just filled with debug stuff
2026-05-22 16:15:36 -07:00
49c2ccd49a
Fixing "dual" modes for Raspberry Pi build, adding logging script
2026-05-22 15:55:37 -07:00
21a6a2f676
creating Pi ready "dual" images ready for public trial
2026-05-22 12:12:07 -07:00
b0b5e4f8cb
Allowing for microReticulum to pair with two Raspberry Pi Zero 2Ws
2026-05-21 17:36:53 -07:00
0c15cf7219
per Codex:jp_native works: jp central -> T-Beam peripheral
...
jp_native_peripheral works: T-Beam central -> jp peripheral
jp_native_dual builds, but needs role arbitration to avoid central/peripheral race
2026-05-21 15:53:31 -07:00
83b122df15
Before testing, but 3rd mode "dual" created for jp
2026-05-21 15:30:46 -07:00
cbc3553436
this role reversal for Intel/jp works, but further work needed to make Intel/jp ambidexterous
2026-05-21 15:09:15 -07:00
0ca0ec605a
build for Intel jp and low grade Bluetooth thereon works and pairs with T-Beam
2026-05-21 14:27:58 -07:00
de75f7d865
306 works, stress tested range between units and learned reconnect logic is needed. Create Exercise 307 for connection resiliency: to be scoped out
2026-05-20 20:23:42 -07:00
5207f72f14
Exercise 305 works great, need to remove or toggle debugging
2026-05-20 17:32:44 -07:00
7410e820c6
All working
2026-05-19 15:38:06 -07:00
31f881233e
Catchup, today has been stampeding horses of productivity
2026-05-19 13:22:39 -07:00
d7258feefe
modified to integrate with Motion_Cal & Magnetometer_Calibration
2026-04-28 13:17:04 -07:00
4397139fb9
Mandatory for environment set up.
2026-04-25 09:43:43 -07:00
b12064755e
Reference to Gentoo build notes
2026-04-25 04:23:28 -07:00
fab25e1a72
Commencing DOxygen documentation
2026-04-25 04:20:14 -07:00
8aff7daa11
Works, but on jp so sluggish as to work intermittently, will try on eos
2026-04-25 04:19:17 -07:00
