Skip to main content

Community

include hvr_stats to alert

Answered

Please sign in to leave a comment.

Comments

2 comments

    Hi Vijayaragavendran,

    Thanks for the request. We already support alerts for things like latency, inactivity, and stalled jobs, but it is not clear yet which specific hvr_stats values the customer wants to alert on.

    hvr_stats covers a broad set of statistics, and turning that into actionable alerting could range from a small extension to a much larger impact. Before we evaluate this further, we should clarify the exact use case. Can you provide that?

    Best regards,

    Edwin

    HI Vijay,

    On alerting: You can configure LatencySla. When a job exceeds the SLA, the SNMP alert sends a trap with the job name—which identifies the channel and location—and the latency in seconds. Since 6.3.5/4, a channel-specific SLA overrides the hub-wide SLA.

    This only alerts on SLA breaches; it doesn’t provide a continuous latency value, and change counts aren’t included in alerting.

    For ongoing monitoring, hvr_stats metrics are available through the REST API endpoint GET /api/v0/hubs/{hub}/stats/metrics. You can filter by channel, location, and metric—for example, Capture Latency Max, Integrate Latency Max, Captured Rows, or Integrated Changes—and poll the endpoint to track those values per channel.

    Would that work?

    Best regards,

    Edwin