SNMP MIB

Prev Next

AppGate ZTNA generates metrics for SNMP and Prometheus using a unified internal method. When the system collects a metric, it formats that metric for both output types simultaneously. As a result, even though Prometheus is the primary output method, a matching metric is almost always available in SNMP.

The SNMP server in AppGate ZTNA exposes metrics as variables organized in a management information base (MIB). The MIB describes the metrics relevant to the software-defined perimeter (SDP) system. Monitoring applications can query these metrics remotely.


Understanding the AppGate MIB

The AppGate ZTNA MIB includes a set of metrics for monitoring the health of appliances in the Collective.

Before you use SNMP, enable the SNMP server in Appliances. You can download the AppGate MIB from Utilities.

The MIB mirrors the Prometheus metrics. The MIB contains four top-level Object Identifiers, each corresponding to the first part of a Prometheus metric name. Under each identifier, the MIB defines a sequence for each metric. For details on each metric, see Prometheus metric details.

Display of Portal DNS Proxy query types, results, and counts in a structured format.

Each sequence contains a list of Object Types. These types map to the labels and values used in Prometheus. For example, the Portal DNS Proxy metric uses two labels—type (a, aaaa) and result (success, cache, nodata, timeout, notfound)—plus the metric value itself. The MIB therefore defines three Object Types for this metric: type, result, and count.

The following sections summarize the key items in the MIB.

sdpApn OBJECT IDENTIFIER ::= { sdp 1 }

Appliance metrics cover CPU usage, memory, disk, network I/O, image size, and status. These metrics overlap with the information displayed in the dashboard.

SPA metrics monitor the front door to appliances from the internet. Use these metrics to confirm that firewalls allow all SPA traffic through. Anomalies in these metrics can also indicate that an appliance is being targeted by a bot attempting to bypass the SPA layer, or that a denial-of-service (DoS) attack is underway.

Proxy_protocol metrics are available when this option is enabled.

SNAT metrics are useful when multiple aliases are in use on the protected network.

audit_logs reports on the logd service running on each appliance, providing visibility into the flow of audit log records. Monitor these metrics to detect interruptions in audit log flow before records are permanently lost or a disk reaches capacity.

audit_events reports the number of log records written for each record type. Use these metrics to monitor specific event counts. For example, you can track the total number of password_authentication_failed events over the previous hour.

sdpGw OBJECT IDENTIFIER ::= { sdp 2 }

xxx_resolver is available in several types: DNS, AWS, and GCP. Slow or error-prone name resolution can degrade system performance. These metrics track DNS queries and API calls using hits, misses, and timeouts, allowing you to identify and resolve name resolution issues quickly.

DNS_forwarder is an additional name resolver type.

Sessions in Gateways relate to access decisions—who gets access to what, and when. The system controls these decisions through events. Monitor these metrics to assess how well Gateways handle their event load. As the number of users on a Gateway increases, this becomes more important because the system schedules events per user.

VPN metrics cover total sessions, firewall rules, and memory usage. These metrics measure Gateway utilization and are typically used to trigger auto-scaling, adding or removing Gateways as needed.

Http reports statistics related to URL Access usage.

Ha provides metrics for high-availability (HA) calls on the protected network relating to Gateway failover. Use these metrics to assess the size, frequency, and timing of failover events.

sdpCtr OBJECT IDENTIFIER ::= { sdp 3 }

Client and Admin metrics cover authentication, authorization, certificate signing requests, Remedy Actions, and device registrations. In addition to operation counts, average processing time is available. When the Controller becomes heavily loaded, processing times increase. Use these metrics to determine when corrective action is needed, such as increasing the number of virtual CPUs (vCPUs).

Ip_pool and License metrics are available.

Database metrics report on the size and state of the database the Controller uses.

Java process monitoring metrics are also available.

sdpPtl OBJECT IDENTIFIER ::= { sdp 4 }

Portal metrics report on the number and state of web clients provisioned on the Portal hosting platform. Monitor these metrics to ensure that available client capacity does not run low.

DNS proxy metrics report on the quantity and results of DNS queries. The DNS proxy, which runs in the Portal, forwards client DNS requests according to each user's DNS settings.