Troubleshooting
11 network domains, advisor mode, and how the assistant diagnoses issues across vendors.
The troubleshooting engine covers 11 network domains across multiple vendors. Describe the problem in plain English, and the assistant walks you through diagnosis using vendor-specific documentation and proven methodology.
Advisor Mode
By default, Damira AI runs in advisor mode. This means:
- The assistant recommends commands for you to run on your devices
- You paste the output back into your AI session
- The assistant analyzes the output and continues diagnosis
- No SSH connections. No credentials stored. Your data stays on your network.
This is the most secure operating mode. The assistant never touches your devices directly.
Supported Domains
| Domain | What It Covers |
|---|---|
| BGP | Peering issues, route advertisement, path selection, community filtering |
| OSPF | Adjacency problems, area configuration, LSA analysis, DR/BDR election |
| NAT | Translation failures, pool exhaustion, PAT conflicts, policy-based NAT |
| MTU | Path MTU discovery, fragmentation issues, black hole detection |
| Switching | VLAN, STP, trunk negotiation, MAC table, port-channel issues |
| Connectivity | Reachability, traceroute analysis, ARP resolution, interface state |
| Voice/UC | SIP registration, one-way audio, call drops, codec negotiation, MRA |
| Wireless | Client association, roaming, channel utilization, AP connectivity |
| Palo Alto | PAN-OS security policies, NAT rules, GlobalProtect, threat logs |
| FortiGate | FortiOS policies, SD-WAN, VPN tunnels, session analysis |
| Junos | SRX/MX/QFX troubleshooting, routing policies, commit issues |
How It Works
- You describe the problem — "OSPF adjacency won't come up between core-rtr-01 and dist-rtr-02"
- Domain identification — The assistant identifies this as an OSPF troubleshooting request
- Evidence you already have — Paste any alerts, metrics, or log excerpts from your monitoring along with the symptoms
- Documentation lookup — Relevant vendor documentation is pulled automatically
- Structured diagnosis — Ranked root causes, with the specific commands to run for anything your evidence doesn't cover
- Iterative analysis — You paste command output, the assistant narrows down the root cause
The assistant follows a structured diagnostic methodology: Gather → Isolate → Diagnose → Recommend → Prevent.
What comes back is a ranked diagnosis, not "check the config":
Start with What Your Monitoring Shows
Paste the firing alert, a metric excerpt, or the relevant log lines from Prometheus, Grafana, Loki, or your syslog server into your AI session along with the symptoms. The assistant uses that evidence first and only asks for CLI output when it needs data your monitoring doesn't capture. See Monitoring.
Tips for Better Results
- Be specific about the problem — "BGP neighbor flapping every 30 seconds" is better than "BGP is broken"
- Include device context — Mention the vendor and platform if possible (e.g., "Cisco IOS-XE 17.x", "Palo Alto PA-5200")
- Paste full command output — The assistant parses structured output from show commands, so include everything
- Use multi-turn conversation — The assistant remembers context within a session. Follow up with additional details as you gather them.
Config Security Audit
Paste a device configuration into your AI session, or point at a file in configs/, with a request like:
The assistant checks for common security gaps: default credentials, unencrypted protocols, missing ACLs, open management interfaces, and more.
The audit runs on your machine, so your config never leaves it. Findings come back by severity (excerpt):
Your AI then gives you the remediation commands for each finding.
Config Generation
Describe what you need and the assistant generates configuration:
- Subnet calculations and IP addressing
- Interface configurations
- Routing protocol setup
- Access control lists
Supported for Cisco, Juniper, Arista, Palo Alto, and Fortinet.