Despite the conf was correct it never worked fine.
With matchers instead of match_re it works like a charm :)
Thanks ;)
Il giorno giovedì 23 giugno 2022 alle 10:54:47 UTC+2 Brian Candler ha
scritto:
> > As far as I know, what you've written is file
>
> s/file/fine/
> On Thursday, 23 June 2022 at 08:10:14 UTC+1 Brian Candler wrote:
>
>> What alertmanager version? Do you have any sort of alertmanager
>> clustering, and if so are you sure the nodes are consistent?
>>
>> As far as I know, what you've written is file. Rules which send to
>> receiver "devops" *should* send to both destinations.
>>
>> Therefore if this isn't happening then you should look at alertmanager
>> logs, to see what it says is happening at delivery time. You should also
>> look at the metrics that alertmanager itself generates; you can scrape
>> these with a separate scrape job, e.g.
>>
>> - job_name: alertmanager
>> scrape_interval: 1m
>> metrics_path: /metrics
>> static_configs:
>> - targets: ['localhost:9093']
>>
>> This will give you counters for delivery attempts and failures for each
>> delivery method, e.g.
>>
>> alertmanager_notification_latency_seconds_bucket{integration="telegram",le="1"}
>>
>> 0
>> alertmanager_notification_latency_seconds_bucket{integration="telegram",le="5"}
>>
>> 0
>> alertmanager_notification_latency_seconds_bucket{integration="telegram",le="10"}
>>
>> 0
>> alertmanager_notification_latency_seconds_bucket{integration="telegram",le="15"}
>>
>> 0
>> alertmanager_notification_latency_seconds_bucket{integration="telegram",le="20"}
>>
>> 0
>> alertmanager_notification_latency_seconds_bucket{integration="telegram",le="+Inf"}
>>
>> 0
>> alertmanager_notification_latency_seconds_sum{integration="telegram"} 0
>> alertmanager_notification_latency_seconds_count{integration="telegram"} 0
>> alertmanager_notification_requests_failed_total{integration="telegram"} 0
>> alertmanager_notification_requests_total{integration="telegram"} 0
>> alertmanager_notifications_failed_total{integration="telegram"} 0
>> alertmanager_notifications_total{integration="telegram"} 0
>>
>> Aside: if you had two different receivers, say "devops_email" and
>> "devops_telegram", then it's possible to write a routing rule which sends
>> to both:
>>
>> # OLD
>> - matchers:
>> - 'foo = bar'
>> receiver: devops
>>
>> # NEW
>> - matchers:
>> - 'foo = bar'
>> routes: [ {receiver: devops_mail, continue: true}, {receiver:
>> devops_telegram} ]
>>
>> However what you wrote originally should work fine as far as I know.
>>
>> On Thursday, 23 June 2022 at 05:05:07 UTC+1 Martin wrote:
>>
>>> Hi there, this is not working for me.
>>> Every notification is sent to either one of the destinations.
>>> I keep receiving or a mail or a telegram message for every firing alert.
>>> Seems to be a kind of round robin. It makes no sense for me.
>>> Any ideas? Have to manage it necessarily at route level?
>>>
>>> Thank you all,
>>>
>>> receivers:
>>> - name: "devops"
>>> email_configs:
>>> - to: "ma...@mail.com"
>>> send_resolved: false
>>> html: '{{ template "mail.template.html" . }}'
>>> headers:
>>> subject: '{{ template "mail.template.subject" . }}'
>>> telegram_configs:
>>> - send_resolved: false
>>> api_url: https://api.telegram.org
>>> bot_token: token.....
>>> chat_id: -123235456.....
>>> message: '{{ template "telegram.template.message" . }}'
>>> disable_notifications: false
>>> parse_mode: ""
>>>
>>
--
You received this message because you are subscribed to the Google Groups
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/prometheus-users/cf55326e-b834-4319-93d7-78a456a0e502n%40googlegroups.com.
