Skip to content

Check type reference

Reference for all check types Vesana supports — sorted by mode.

Active checks (from server)

Run on the Vesana server itself. Mostly for self-monitoring; for customer hosts use a collector.

ping

ICMP echo to target.

Config Default Meaning
count 4 Packet count
timeout_ms 1000 Timeout per packet

Result: OK if count succeed, WARNING on packet loss, CRITICAL if all lost.

check_type: ping
check_config:
  count: 4
  timeout_ms: 1000

http

HTTP/HTTPS request with optional body validation.

Config Default Meaning
url – URL incl. scheme
method GET HTTP method
expected_status 200 Expected status code
body_contains – Substring in response body
timeout_s 10 Connection timeout
verify_tls true Verify TLS cert 
check_type: http
check_config:
  url: https://api.example.com/health
  expected_status: 200
  body_contains: '"ok":true'

port

TCP connect test.

Config Default Meaning
port – TCP port
timeout_s 5 Connect timeout

ssl_certificate

Certificate expiry check.

Config Default Meaning
host – Hostname
port 443 TLS port
warn_days 21 WARN at < N days remaining
crit_days 7 CRITICAL at < N days

Passive checks (from collector)

Run by a collector in the customer network. Prerequisite: host.collector_id set.

snmp

Single OID, numeric comparison.

Config Default Meaning
oid – numeric OID
version 2c 1, 2c, 3
community – community (encrypted)
multiplier 1 value × multiplier 
check_type: snmp
check_config:
  oid: ".1.3.6.1.4.1.9.9.13.1.3.1.3.1"
  version: 2c
  community: public
threshold_warn: 60
threshold_crit: 75
value_type: gauge

snmp_string

OID, value as string, optional pattern match.

Config Meaning
oid OID
match_value expected value or regex
mode exact, regex, contains

snmp_table

Bulk walk on a table, result as JSONB in check_results.perfdata. Foundation for switch_portmap and nas_diskbay visuals.

check_type: snmp_table
check_config:
  walk_oid: ".1.3.6.1.2.1.2.2.1"
  columns:
    name:        ".1.3.6.1.2.1.2.2.1.2"
    speed:       ".1.3.6.1.2.1.2.2.1.5"
    oper_status: ".1.3.6.1.2.1.2.2.1.8"

ssh_command

SSH login, run command, evaluate exit code + output.

Config Meaning
username SSH user
auth_method password or key
password (encrypted)
private_key PEM content (encrypted)
command Shell command
expected_exit_code Default 0

ssh_cpu, ssh_mem, ssh_disk

Wrappers around ssh_command with standard commands — typical for ESXi and agentless Linux.

http and port (passive variant)

Same config as active, but run by collector — sees target from customer network, not from Vesana server.

Agent checks

Prerequisite: host.agent_managed = true, agent installed, valid token.

agent_cpu

CPU usage in %, averaged over the last interval.

check_type: agent_cpu
threshold_warn: 80
threshold_crit: 95
value_type: gauge

agent_memory

RAM usage in % (Used / Total).

agent_disk

Disk usage per mount in %.

Config Meaning
path /, /var, C:, D:

agent_service

OS service status.

Config Meaning
service systemd unit or Windows service name

Result: OK when active, CRITICAL otherwise.

agent_process

Are at least N processes running with this name?

Config Meaning
process Process name or pattern
min_count Default 1
max_count optional

agent_eventlog (Windows)

Evaluate Windows event log.

Config Meaning
log Application / System / Security
level Error / Warning / Information
minutes Lookback window
source optional source filter

Result: count of entries in window.

agent_custom

Own command with output pattern matching.

Config Meaning
command Shell command
ok_pattern Regex for OK
warn_pattern Regex for WARNING
crit_pattern Regex for CRITICAL
timeout_s Default 30 
check_type: agent_custom
check_config:
  command: "/usr/local/bin/checkbackup.sh"
  ok_pattern: "^Backup OK"
  crit_pattern: "(failed|error)"

agent_script

Run server-managed monitoring script. See Monitoring scripts.

Config Meaning
script_id Script UUID

agent_services_auto (Windows)

Check all auto-start services — any not running is CRIT.

Config Meaning
exclude List of service names to ignore

Special

nsca_passive

Packets from the NSCA receiver land as „virtual checks" with check_mode = passive in the pipeline. Detail: NSCA migration.

Common fields (all check types)

Field Meaning
interval_seconds how often (10 s ≤ x ≤ 7 days)
retry_interval_seconds in soft state (≥ 5 s, ≤ main interval)
max_check_attempts soft → hard after N failures
threshold_warn numeric warn threshold
threshold_crit numeric crit threshold
value_type gauge / counter / status / info / duration

Result format

Every result reaching the receiver has:

{
  "host_token_or_api_key": "...",
  "service_id": "uuid",
  "status": 0,           // 0 OK, 1 WARN, 2 CRIT, 3 NO_DATA, 4 UNKNOWN
  "value": 42.5,
  "message": "OK - load 42.5%",
  "perfdata": { "load1": 0.42, "load5": 0.31 },
  "checked_at": "2026-04-25T10:15:30Z"
}

perfdata (JSONB) is optional — used for charts and visuals.

Which check type for which problem

Problem Recommendation
Is the server reachable? ping (passive)
Is the website up? http (passive)
Is Apache running? agent_service (Linux/Windows)
CPU load? agent_cpu
Disk filling up? agent_disk with path
Switch port status? snmp_table with ifTable, visual switch_portmap
UPS battery voltage? snmp with OID from UPS-MIB
Backup yesterday OK? agent_script with custom script parsing log
Own logic? agent_custom with pattern matching
ESXi without agent? ssh_cpu / ssh_mem / ssh_disk

Next