Hi,
According to log I'm using UDP subscriptions (default kapacitor config).
Here's are the details of one of the deadman tasks:
=============
ID: dead_agg_load
Error:
Template:
Type: stream
Status: disabled
Executng: false
Created: 15 Jun 16 12:53 UTC
Modified: 16 Jun 16 15:09 UTC
LastEnabled: 16 Jun 16 13:17 UTC
Databases Retenton Policies: ["metrics"."default"]
TICKscript:
stream
|from()
.measurement('duration')
.where(lambda: "name" == 'agg_load')
|eval(lambda: "value")
.as('value')
|deadman(1.0, 5m)
.message('service: Agg Load is {{ if eq .Level "OK" }}alive{{ else
}}dead - not running for 2h{{
end }}')
.log()
.slack()
.channel('#alerts')
DOT:
digraph dead_agg_load {
stream0 -> from1;
from1 -> eval2;
eval2 -> noop4;
stats3 -> derivative5;
derivative5 -> alert6;
}
===============
Thanks!
BR,
Elad
On Thursday, June 16, 2016 at 9:17:56 PM UTC+3, [email protected] wrote:
>
> Some basic trouble shooting steps/questions
>
> 1. How are you sending data to Kapacitor? Over UDP subcriptions? Is it
> possible that while InfluxDB has the data it was dropped before it made it
> to Kapacitor?
> 2. Can you share the `kapacitor show` output of your task?
>
>
>
> On Thursday, June 16, 2016 at 5:12:46 AM UTC-6, [email protected] wrote:
>>
>> yes? no? maybe?
>> someone?
>>
>>
>> On Wednesday, June 15, 2016 at 6:47:46 PM UTC+3, [email protected] wrote:
>>>
>>> For reference -
>>> Here's the *DERIVATIVE* example I've configured, but still sending
>>> alerts for existing/new points:
>>> -------------
>>> stream
>>> |from()
>>> .measurement('delaware:last_completed')
>>> .where(lambda: "name" == 'exchange_hourly')
>>> |stats(1m)
>>> |derivative('emitted')
>>> .unit(1m)
>>> .nonNegative()
>>> |alert()
>>> .id('{{ .TaskName }}')
>>> .message('service is dead')
>>> .crit(lambda: "emitted" < 1.0)
>>> .log()
>>> .slack()
>>> .channel('#alerts')
>>> -----------------
>>> (Also tried to implement this derivative within a batch script. same
>>> behavior)
>>>
>>> Please advise, as many of our services depends on this feature :/
>>> thanks a lot.
>>>
>>> On Wednesday, June 15, 2016 at 5:42:54 PM UTC+3, [email protected]
>>> wrote:
>>>>
>>>> Hello,
>>>>
>>>> We're trying to setup a *deadman* check for two measurements,
>>>> but after long hours and QA to this function - I cannot get rid of the
>>>> false-positives it sends.
>>>>
>>>> Could you please take a look at this TICK and let me know what am I
>>>> doing wrong?
>>>> Here's the TICK file:
>>>>
>>>> stream
>>>> |from()
>>>> .measurement('last_completed')
>>>> .where(lambda: "name" == 'data')
>>>> // |eval(lambda: "value").as('value')
>>>> |deadman(1.0, 1m)
>>>> .message('Service:{{ index .Tags "name"}} is {{ if eq .Level "OK"
>>>> }}alive{{ else }}dead - not running for 1m {{ end }}')
>>>> .log()
>>>> .slack()
>>>> .channel('#alerts')
>>>>
>>>>
>>>> And here's an example data set:
>>>>
>>>>
>>>> 2016-06-15T14:05:47.769283954Z "data" 1
>>>> 2016-06-15T14:05:56.17229738Z "data" 1
>>>> 2016-06-15T14:06:04.216883312Z "data" 3
>>>> 2016-06-15T14:06:12.028630147Z "data" 2
>>>> 2016-06-15T14:06:20.21923461Z "data" 2
>>>> 2016-06-15T14:06:28.37728243Z "data" 0
>>>> 2016-06-15T14:06:36.239360137Z "data" 1
>>>> Although there are new points every 8 seconds approx, deadman is
>>>> alerting every minute.
>>>> I've tried all kind of times & limits.. (checks for every minute, hour,
>>>> two hours).
>>>>
>>>> What could be the reason for this?
>>>>
>>>> Important --
>>>> I've also tried to convert the above STREAM tick to a BATCH and the
>>>> behavior was the same.
>>>> Also tried to use the DERIVATIVE instead of the deadman - and it still
>>>> ignores the new points and alerting :(
>>>>
>>>>
>>>> Influx 0.13.0
>>>> Kapacitor 1.0.0-beta
>>>> (the above was tried on both 0.13.1 and the new beta)
>>>>
>>>> THANKS!
>>>> Elad
>>>>
>>>
--
Remember to include the InfluxDB version number with all issue reports
---
You received this message because you are subscribed to the Google Groups
"InfluxDB" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/influxdb.
To view this discussion on the web visit
https://groups.google.com/d/msgid/influxdb/cf5e80cb-5610-45eb-9d1d-e9cdcc6df0d1%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
