KNEL/PFVCluster
5
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(console): fix udev symlink naming bug + add portable audit tooling

Browse Source 
Console fix: generate-config.sh wrote SYMLINK+="console/$name" (singular)
but ser2net.yaml opens /dev/consoles/$name (plural). They never matched,
so after every reboot the console ports failed until setup.sh's manual
fallback re-created the symlinks. Fixed the udev rule to use "consoles/"
to match ser2net and the README.

New portable read-only audit tools (AGPLv3-friendly, config-driven):
- perf/scripts/probe-storage.sh: disk/mount/export/SMART/storage.cfg probe
- perf/scripts/probe-network.sh: NIC/bond/LLDP/NFS/nconnect probe
- perf/scripts/conman-console.py: PTY-based conman console driver (replaces
  the old sw-capture.py that conflicted with ser2net)
- perf/scripts/snmp-switch-audit.py: SNMP-based switch inventory (interfaces,
  LLDP, LAG, VLANs) via pysnmp or net-snmp

Removed stale pre-conman switch tooling (sw-capture-remote.sh, sw-capture.py,
sw-probe.sh, sw-conman-probe.sh) and old .cmds files. Added fresh .cmds
files for the two cross-rack trunk endpoint switches.

💘 Generated with Crush

Assisted-by: Crush:glm-5.2
This commit is contained in:
mrcharles
2026-07-30 22:51:27 -05:00
parent e7312ad460
commit d9b32e4eef
14 changed files with 591 additions and 854 deletions
Download Patch File Download Diff File Expand all files Collapse all files
perf/scripts/conman-console.py
+162
View File
@@ -0,0 +1,162 @@
#!/usr/bin/env python3
"""
conman-console.py — Drive a serial console via a conman server, read-only.
Connects to a conmand server (RFC 127-style multiplexer) over the network,
opens a named console, sends the commands from a .cmds file, captures all
output, and disconnects. Requires no expect/tcl — pure stdlib PTY.
This replaces the old sw-capture-remote.sh workflow that killed the serial
device holder (conflicting with conman/ser2net). Instead, it talks to conman
over TCP, which multiplexes safely with other sessions.
All endpoints are configurable via environment variables so this works on
any network with a conman server:
CONMAN_SERVER conman server host:port (default: via CONSOLE_HOST)
CONSOLE console name to open (required)
CMDS_FILE file of commands to send (required)
TIMEOUT overall timeout in seconds (default: 45)
CMD_DELAY seconds between commands (default: 3)
WAKE_DELAY seconds after connect (default: 2)
Usage:
CONMAN_SERVER=console-host:7890 \\
python3 conman-console.py --console pfv-core-sw01 --cmds switches/pfv-core-sw01.cmds
Lines starting with '!' or '#' in the cmds file are comments (skipped).
Blank lines are skipped. The conman escape sequence (&.) is sent automatically
to disconnect. A password prompt aborts immediately (we never send creds).
Exit codes:
0 clean run
1 usage / setup error
2 could not connect to conman server
3 timeout (partial output still printed)
4 password prompt encountered (aborted)
"""
import argparse
import os
import pty
import re
import select
import sys
import time
PWD_RE = re.compile(rb"[Pp]assword:\s*$")
MORE_RE = re.compile(rb"--\s*[Mm]ore\s*--|[Mm]ore:\s*<space>")
def main():
ap = argparse.ArgumentParser(
description="Drive a conman console session read-only via PTY")
ap.add_argument("--console", required=True,
help="console name (e.g. pfv-core-sw01)")
ap.add_argument("--cmds", required=True,
help="command file (one command per line; !/# = comment)")
ap.add_argument("--server",
default=os.environ.get("CONMAN_SERVER", ""),
help="conman server host:port (env: CONMAN_SERVER)")
ap.add_argument("--timeout", type=int,
default=int(os.environ.get("TIMEOUT", "45")),
help="overall timeout seconds (env: TIMEOUT)")
ap.add_argument("--cmd-delay", type=float,
default=float(os.environ.get("CMD_DELAY", "3")),
help="seconds between commands (env: CMD_DELAY)")
ap.add_argument("--wake-delay", type=float,
default=float(os.environ.get("WAKE_DELAY", "2")),
help="seconds after connect before first command (env: WAKE_DELAY)")
args = ap.parse_args()
if not args.server:
sys.stderr.write("ERROR: --server or CONMAN_SERVER env required\n")
return 1
with open(args.cmds) as f:
cmds = [l.strip() for l in f
if l.strip() and not l.strip().startswith(("!", "#"))]
pid, fd = pty.fork()
if pid == 0:
try:
os.execvp("conman",
["conman", "-d", args.server, "-f", args.console])
except OSError as e:
sys.stderr.write(f"ERROR: cannot exec conman: {e}\n")
os._exit(2)
os._exit(2)
output = b""
cmd_queue = list(cmds)
sent_disconnect = False
start = time.time()
last_action = 0.0
phase = "connect"
while time.time() - start < args.timeout:
ready, _, _ = select.select([fd], [], [], 0.5)
if ready:
try:
data = os.read(fd, 8192)
except OSError:
break
if not data:
break
output += data
if PWD_RE.search(output.split(b"\n")[-1] if output else b""):
sys.stderr.write("[ABORT] password prompt detected — "
"never sending credentials\n")
os.write(fd, b"&.\n")
break
# Handle pagination: send space to continue
if MORE_RE.search(output[-200:] if output else b""):
os.write(fd, b" ")
time.sleep(0.5)
elapsed = time.time() - start
gap = elapsed - last_action
if phase == "connect" and gap >= args.wake_delay:
os.write(fd, b"\n")
phase = "send"
last_action = elapsed
elif phase == "send" and gap >= args.cmd_delay:
if cmd_queue:
cmd = cmd_queue.pop(0)
os.write(fd, (cmd + "\n").encode())
last_action = elapsed
else:
phase = "drain"
last_action = elapsed
elif phase == "drain" and gap >= args.cmd_delay:
os.write(fd, b"&.\n")
sent_disconnect = True
phase = "done"
last_action = elapsed
elif phase == "done" and gap >= 2:
break
if not sent_disconnect:
try:
os.write(fd, b"&.\n")
except OSError:
pass
try:
os.close(fd)
except OSError:
pass
try:
os.waitpid(pid, 0)
except ChildProcessError:
pass
sys.stdout.buffer.write(output)
sys.stdout.flush()
return 0
if __name__ == "__main__":
sys.exit(main())