serversdown/seismo-relay
1
0
Fork 0
Code Issues 4 Pull Requests Actions Packages Projects Releases Wiki Activity

fix(protocol): remove auto-detection of frame mode and ensure extra chunks are always used for valid waveform footer

Browse Source
This commit is contained in:
Brian Harrison
2026-04-28 00:05:51 -04:00
parent 6b875e161b
commit e1a6fd5386
2 changed files with 24 additions and 29 deletions
Download Patch File Download Diff File Expand all files Collapse all files
CLAUDE.md
+18 -6
View File
@@ -350,14 +350,20 @@ Do NOT use fixed absolute offsets for sample_rate or record_time.
**SUB 5A (bulk waveform) TCP frame splitting — confirmed 2026-04-27:**
Over TCP via cellular modem, each 5A chunk request that produces a single ~1100-byte
A5 response over direct RS-232 arrives as **two separate, complete S3 frames** of
~550 bytes each. This is because the device sends its RS-232 response as multiple
frames, and the modem's Data Forwarding Timeout (~100-150 ms) delivers them to us
as separate TCP segments, each parsed as a complete S3 frame.
A5 response over direct RS-232 may arrive as **two separate, complete S3 frames** of
~550 bytes each ("2-frame mode"). The modem's Data Forwarding Timeout (~100-150 ms)
can split the RS-232 response into two TCP segments, each parsed as a complete S3 frame.
Under different modem/timing conditions the full ~1100-byte response arrives as **one
S3 frame** ("1-frame mode").
**Both modes require `extra_chunks_after_metadata=1`** (the extra chunk at metadata_counter
+ 0x0400). The device's waveform footer data lives at circular-buffer address 0x1C00 for
this event; the terminator frame must be sent at 0x1C00 (not 0x1800) to receive it.
Example for a 2-second Continuous event (BE11529, key=01110000) via TCP:
- 1 probe frame (554 B) + 5 chunks × 2 frames (556-573 B) + 1 extra chunk × 2 frames + 1 terminator (208 B) = **14 A5 frames**
- All 14 frames contribute body data; using all of them gives the correct 6864-byte file.
- **2-frame mode:** 1 probe frame (554 B) + 5 chunks × 2 frames (556-573 B) + 1 extra chunk × 2 frames + 1 terminator (208 B) = **14 A5 frames** → 6864-byte file
- **1-frame mode:** 1 probe frame (~1097 B) + 5 chunks × 1 frame (~1079-1113 B) + 1 extra chunk × 1 frame (smaller, tail of event) + 1 terminator → **8 A5 frames** → 6864-byte file
- All frames contribute body data; using all of them gives the correct file.
**Fix (confirmed 2026-04-27):** `_recv_5a_batch()` in `protocol.py` collects ALL
A5 frames per chunk request before the next request is sent, using a 0.5 s batch
@@ -365,6 +371,12 @@ timeout after the first frame to catch the ~150 ms delayed second frame. `write
includes ALL body frames without skipping — the extra chunk's frames are part of the
body data, NOT padding to be discarded.
**WRONG earlier hypothesis (do not re-introduce):** An attempt was made to auto-detect
1-frame vs 2-frame mode from the probe frame size and skip the extra chunk when
`probe_data_len >= 700`. This was wrong — the extra chunk is always needed to advance
the device's internal state to the footer address. The `_probe_is_large` branch was
removed 2026-04-27.
### Required ACEmanager settings (Sierra Wireless RV50/RV55)
| Setting | Value | Why |