trishul-snmp-suite

Troubleshooting

This guide covers the most common 2.0.2 runtime and operator issues.

The App Does Not Start

Check:

Common causes:

If needed, choose different ports:

APP_PORT=9080 BACKEND_PORT=9000 SNMP_PORT=2161 TRAP_PORT=2162 ./install-trishul-snmp-suite.sh up

The UI Loads But Login Fails

Check:

Important current-line note:

If you lost access, recover from a backup or edit the SQLite-backed auth records offline before restarting the app.

MIB Manager Upload Or Reload Fails

Check:

If dependency fetch is disabled in your environment, upload missing MIBs manually first.

Symbolic Names Do Not Resolve

Check:

If no matching MIBs are loaded, use numeric OIDs until the catalog is ready.

Simulator Will Not Start

Common causes:

For same-host container validation, explicitly use the UDP ports exposed by your deployment, such as 1061.

Traps Listener Starts But No Events Appear

Check:

If events still do not appear, restart the listener and repeat the test with a simple loopback target such as 127.0.0.1:1162.

Walk & Parse Fails

Check:

When validating locally in Docker, prefer the exposed host UDP port instead of the form defaults if those differ.

Upgrade From 1.x Looks Empty

This can happen even when the old Docker volume was copied forward.

The new runtime uses SQLite and compiled bundle artifacts, so validate:

  1. MIBs in MIB Manager
  2. simulator custom data in Simulator
  3. traps flow in Traps
  4. metadata and settings in Dashboard and Settings

Where Runtime State Lives

Default container path:

Important runtime paths:

Container logs are not persisted in the data volume by default. Read them with ./install-trishul-snmp-suite.sh logs or docker logs trishul-snmp-suite. /app/backend/data/logs/backend.log is only used when file logging is explicitly enabled.

Useful Commands

./install-trishul-snmp-suite.sh status
./install-trishul-snmp-suite.sh logs
npm --prefix frontend run build
.venv/bin/python -m pytest backend/ -q
.venv/bin/python scripts/check_backend_coverage.py

If the problem is still unclear, reduce the workflow to:

  1. log in
  2. start a responder in Simulator
  3. run one walk in Walk & Parse
  4. start a listener in Traps
  5. send one local trap