Alert data outputs to children nodes

[nathanielc]

Fixes #93

Alert data can now be consumed directly from within TICKscripts.
For example, let's say we want to store all data that triggered an alert in InfluxDB with a tag level containing the level string value (i.e CRITICAL).

...
    |alert()
        .warn(...)
        .crit(...)
        .levelTag('level')
         // Or use a field
        .levelField('level')
    |influxDBOut()
        .database('alerts')
        ...

[lpalm]

Does levelTag() work for deadman alerts? I tried this but the points being inserted contain only tags for each of the groupBy tags and the one value field for each point. No "level" tag is attached to the points inserted into InfluxDB.

Kapacitor 1.0.0~beta1 (git: master c6f7db5)

var b = stream
    |from()
        .database('X')
        .retentionPolicy('autogen')
        .measurement('X')
        .where(...)
        .groupBy('X', 'Y', 'Z')

b
    |deadman(1.0, 1m)
        .id('streamingDeadman {{ .Group }}')
        .message('{{ .ID }} is {{ .Level }}: {{ index .Fields "emitted" | printf "%.0f" }} points seen in last 1m')
        .levelTag('level')
        .stateChangesOnly()
        .log('X')
        .mode(644)
        .post('X')

b
    |influxDBOut()
        .database('X')
        .retentionPolicy('autogen')
        .measurement('streamingDeadman')

[nathanielc]

It does but this script is not sending the deadman alert to InfluxDB but rather the original data selected from the stream

Did you mean this?

var b = stream
    |from()
        .database('X')
        .retentionPolicy('autogen')
        .measurement('X')
        .where(...)
        .groupBy('X', 'Y', 'Z')

b
    |deadman(1.0, 1m)
        .id('streamingDeadman {{ .Group }}')
        .message('{{ .ID }} is {{ .Level }}: {{ index .Fields "emitted" | printf "%.0f" }} points seen in last 1m')
        .levelTag('level')
        .stateChangesOnly()
        .log('X')
        .mode(644)
        .post('X')
    |influxDBOut()
        .database('X')
        .retentionPolicy('autogen')
        .measurement('streamingDeadman')

[lpalm]

That's exactly right, @nathanielc. Confirmed this works. Thanks and sorry to waste your time with that.

While we're here, what would be the easiest way to add a field to the inserted point that's equal to the alert's message?

[nathanielc]

@lpalm That is currently not possible without custom code like a UDF. But it is a good idea and would follow the work from this PR well. Feel free to create an issue to track the request. PRs are always welcome ;)

[manojgour]

Following below doc for writing stream or batch data back to Influx:
https://docs.influxdata.com/kapacitor/v1.1/examples/continuous_queries/

Doesn't seem to be working for me; Here's my TICK script:

kapacitor show CQ_engine
ID: CQ_engine
Error:
Template:
Type: batch
Status: enabled
Executing: true
Created: 23 Jan 17 13:50 EST
Modified: 23 Jan 17 13:53 EST
LastEnabled: 23 Jan 17 13:53 EST
Databases Retention Policies: ["kapacitor_example"."autogen"]
TICKscript:
batch
|query('SELECT mean(usage_idle) as usage_idle FROM "kapacitor_example"."autogen".cpu')
.period(5m)
.every(5m)
.groupBy(*)
|influxDBOut()
.database('kapacitor_example')
.retentionPolicy('autogen')
.measurement('mean_cpu_idle')
.precision('s')

DOT:
digraph CQ_engine {
graph [throughput="0.00 batches/s"];

query1 [avg_exec_time_ns="0s" batches_queried="0" connect_errors="0" points_queried="0" query_errors="0" ];
query1 -> influxdb_out2 [processed="0"];

influxdb_out2 [avg_exec_time_ns="0s" points_written="0" write_errors="0" ];
}

Also tried working with stream TICK script:

kapacitor show CQ_engine_stream
ID: CQ_engine_stream
Error:
Template:
Type: stream
Status: enabled
Executing: true
Created: 23 Jan 17 14:01 EST
Modified: 23 Jan 17 14:01 EST
LastEnabled: 23 Jan 17 14:01 EST
Databases Retention Policies: ["kapacitor_example"."autogen"]
TICKscript:
stream
|from()
.database('kapacitor_example')
.measurement('cpu')
.groupBy(*)
|window()
.period(5m)
.every(5m)
.align()
|mean('usage_idle')
.as('usage_idle')
|influxDBOut()
.database('kapacitor_example')
.retentionPolicy('autogen')
.measurement('mean_cpu_idle')
.precision('s')

DOT:
digraph CQ_engine_stream {
graph [throughput="0.00 points/s"];

stream0 [avg_exec_time_ns="0s" ];
stream0 -> from1 [processed="0"];

from1 [avg_exec_time_ns="0s" ];
from1 -> window2 [processed="0"];

window2 [avg_exec_time_ns="0s" ];
window2 -> mean3 [processed="0"];

mean3 [avg_exec_time_ns="0s" ];
mean3 -> influxdb_out4 [processed="0"];

influxdb_out4 [avg_exec_time_ns="0s" points_written="0" write_errors="0" ];
}

Appreicate your help here.

[nathanielc]

@manojgour We try to reserve GitHub Issues for actionable bug reports or feature requests. Please send general questions to the [email protected] mailing list.