retention piece 1: immediate rotations for chat-history / job-log / followups (b67084660385)

- chat-history.jsonl rotates at 10MB or 7d -> logs/archive/*.jsonl.gz + .sha256
- job-log.jsonl rotates at 2MB or 14d -> same archive layout
- followups.json archives resolved>7d / escalated>30d -> followups.archive.jsonl (append-only)
- verify wrapper: sha256 -c, gzip -t, clean listing, spot-extract JSON + ts bounds
- driver + hourly systemd user timer (retention-rotations.timer)
- archive-only: nothing is ever deleted; thread backfill excluded (needs human-reviewed preview)
First run 2026-10-05 16:35Z: chat-history 29.7MB + job-log 4.7MB rotated and verified; followups 0 eligible of 163.

bin/retention-archive-followups.py

@@ -0,0 +1,93 @@
+#!/usr/bin/env python3
+"""retention-archive-followups.py — retention piece 1 (board bug b67084660385).
+
+Archives resolved follow-ups older than 7 days and escalated follow-ups older
+than 30 days from followups.json into followups.archive.jsonl (append-only).
+
+ARCHIVE-ONLY: records are moved from the live file to the archive file; the
+archive itself is never pruned or hard-deleted by this script.
+
+Concurrency: the live file is written atomically (tmp + os.replace) by both
+this script and bin/followup-sweeper.py. This script snapshots the file's
+(mtime_ns, size) before deciding the archival set and aborts (rc=2) if the
+file changed in the meantime — the next run picks it up.
+"""
+import json
+import os
+import sys
+from datetime import datetime, timezone, timedelta
+
+ROOT = os.environ.get("NETVM_ROOT", "/home/super/Projects/NetVM")
+LIVE = os.path.join(ROOT, "followups.json")
+ARCHIVE = os.path.join(ROOT, "followups.archive.jsonl")
+RESOLVED_AFTER_DAYS = 7
+ESCALATED_AFTER_DAYS = 30
+
+
+def parse_ts(ts):
+    if not ts:
+        return None
+    try:
+        dt = datetime.fromisoformat(str(ts).replace("Z", "+00:00"))
+        return dt if dt.tzinfo is not None else dt.replace(tzinfo=timezone.utc)
+    except Exception:
+        return None
+
+
+def main():
+    if not os.path.exists(LIVE):
+        print("SKIP followups: no live file at %s" % LIVE)
+        return 0
+
+    with open(LIVE, encoding="utf-8") as f:
+        raw = f.read()
+    st_before = os.stat(LIVE)
+    data = json.loads(raw)
+    now = datetime.now(timezone.utc)
+
+    archive, keep = [], {}
+    for key, rec in data.items():
+        status = rec.get("status")
+        eligible = False
+        reason = ""
+        if status == "resolved":
+            dt = parse_ts(rec.get("resolved_at")) or parse_ts(rec.get("sent_at"))
+            if dt and (now - dt) > timedelta(days=RESOLVED_AFTER_DAYS):
+                eligible, reason = True, "resolved>7d"
+        elif status == "escalated":
+            dt = parse_ts(rec.get("escalated_at")) or parse_ts(rec.get("sent_at"))
+            if dt and (now - dt) > timedelta(days=ESCALATED_AFTER_DAYS):
+                eligible, reason = True, "escalated>30d"
+        if eligible:
+            out = dict(rec)
+            out["_archived_at"] = now.isoformat()
+            out["_archive_reason"] = reason
+            archive.append(out)
+        else:
+            keep[key] = rec
+
+    if not archive:
+        print("SKIP followups: 0 eligible of %d records (resolved>7d / escalated>30d)" % len(data))
+        return 0
+
+    st_after = os.stat(LIVE)
+    if (st_after.st_mtime_ns, st_after.st_size) != (st_before.st_mtime_ns, st_before.st_size):
+        print("ABORT followups: live file changed during archival decision; retry next run",
+              file=sys.stderr)
+        return 2
+
+    with open(ARCHIVE, "a", encoding="utf-8") as f:
+        for rec in archive:
+            f.write(json.dumps(rec) + "\n")
+
+    tmp = "%s.tmp.retention.%d" % (LIVE, os.getpid())
+    with open(tmp, "w", encoding="utf-8") as f:
+        json.dump(keep, f, indent=2)
+    os.replace(tmp, LIVE)
+
+    print("OK followups: archived %d records -> %s (live now %d)" % (len(archive), ARCHIVE, len(keep)))
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())