hansjone/netx
1
0
Fork 0
mirror of https://github.com/hansjone/netx.git synced 2026-10-09 11:50:44 +08:00
Code Issues Projects Releases 1 Packages Wiki Activity Actions 2

Harden long biz_state collects: release DB during CLI and spool before flush.

Browse source 
Avoid idle Session disconnect after heavy timeout, and separate SSH collect from batched Postgres persist via disk spool.

Co-authored-by: Cursor <cursoragent@cursor.com>
This commit is contained in:
oliver 2026-09-22 16:24:46 +08:00
parent c15ea00a23
commit 0eb65a839e
5 changed files with 1293 additions and 433 deletions
Download patch file Download diff file Expand all files Collapse all files

5
netx_api/config.py
View file

@ -122,6 +122,11 @@ class Settings(BaseSettings):
biz_state_heavy_read_timeout_sec: int = 1500
biz_state_heavy_run_timeout_cap_sec: int = 2400
biz_state_heavy_workers: int = 4
# Collect spool: CLI/raw + parsed records on disk; flush to DB every N cmds.
biz_state_spool_dir: str = "data/biz_state_spool"
biz_state_persist_every_cmds: int = 8
# Cap raw_text loaded into Postgres from spool (0 = unlimited).
biz_state_raw_max_bytes: int = 8 * 1024 * 1024
# Managed NE exec: max CLI commands per request (lab can raise; hard-capped in ne_exec).
ne_exec_max_commands: int = 5
# Opt-in: allow per-NE exec_policy (linux_shell/unrestricted). Default off — UI hidden.