[prometheus-users] Pushing 3rd party VNF metrics into node exporter

Hi, Is there any way to push 3rd party VNF metric into the node exporter and also I am not able to see LLC-stores metrics when the data is exported via node exporter with collector.perf option enabled. Thanks, Divya.R -- You received this message because you are subscribed to the Google G

[prometheus-users] How do I create labeled metrics with different labels in a threaded environment?

I have an application that spawns many worker threads. In the main application I create a PrometheusMetrics() helper class that registers the metrics the threads will report on with the appropriate label names. I pass the PrometheusMetrics class into each thread, and then create local variabl

[prometheus-users] Fwd: [prometheus-developers] [RFC][Proposal]: Alertmanager Time Intervals

Hi everyone, Thank you all for your feedback on the Alertmanager Time Interval design doc so far. The design has been greatly simplified and (in my opinion) improved already due to the feedback received. I've made some changes to the proposed implementation so that it works on receivers now, not

[prometheus-users] Windows Metrics

Can I have scripts/Metrics for Windows clients in Grafana § CPU utilization § File System utilization § System messages (ex. Errors in /var/log/messages) Any critical errors reported in logs § Disk INODE utilization

Re: [prometheus-users] Metrics timeseries lag on master prometheus of federated metrics

On 09.06.20 23:23, Raghavendra k wrote: > > With two different hosted solutions Azure (for applications) & AWS (for > centralized monitoring), we are trying to federate the Prometheus running on > the Azure cluster from the Prometheus running on the AWS cluster. > Federation is working absolutely

Re: [prometheus-users] Preventing data loss from poor network communication

Alright, thank you for your time. On Mon, Jun 15, 2020 at 2:19 PM Stuart Clark wrote: > The Push Gateway isn't a caching system. If the Prometheus server can't > connect to fetch a scrape due to network issues you will miss data. The > server needs to have reliable connectivity to the systems it

Re: [prometheus-users] Issues sending alerts through SendGrid

Ok, i changed the log level to debug and I'm seeing that its "sending" the notifications: level=debug ts=2020-06-15T18:17:54.150Z caller=dispatch.go:465 component=dispatcher aggrGroup="{}:{alertname=\"InstanceDown\"}" msg=flushing alerts=[InstanceDown[48afd0d][active]] level=debug ts=2020-0

Re: [prometheus-users] Preventing data loss from poor network communication

The Push Gateway isn't a caching system. If the Prometheus server can't connect to fetch a scrape due to network issues you will miss data. The server needs to have reliable connectivity to the systems it is scraping. On 15 June 2020 13:29:20 BST, "Mathieu Tétreault" wrote: >Alright, I'll loo

Re: [prometheus-users] Missed clients

On 15 Jun 08:46, 'Metrics Searcher' via Prometheus Users wrote: > Hello. > > Could you please give me a clue how to find lost clients from the OpenVPN > server. The metric looks like below: > openvpn_server_client_received_bytes_total{common_name="ID",connection_time="1591021598",instance="server

[prometheus-users] Missed clients

Hello. Could you please give me a clue how to find lost clients from the OpenVPN server. The metric looks like below: openvpn_server_client_received_bytes_total{common_name="ID",connection_time="1591021598",instance="server.com:9176",job="server.com",real_address="1.1.1.1:56228",status_path="/etc

[prometheus-users] Re: How to define exact hostname or IP in prometheus.yml file

Hi Brian, Can you give example how to do set__metric__path since the format of alert which we wanted was achieved , but the end point is still showing down in prometheus., as its not able to scrape the metrics due to metrics path not correct. http://10.2.3.17:8080/metrics in this instead o

[prometheus-users] Re: How to define exact hostname or IP in prometheus.yml file

I have done the changes as suggested and it worked. Thanks a lot for your help. Below is the format we are receiving the alert now. *Labels* alertname = InstanceDown customer = APIGEE instance = 10.2.0.85 job = apigeenode severity = critical *Annotations* description = 10.2.0.85 of job apigeenode

Re: [prometheus-users] Re: Last value(metric count) of a range vector

On Mon, 15 Jun 2020 at 12:41, Baitanik Talukder wrote: > Yeah it is pushed to pushgateway.. > so here for last 4 days no metric was pushed to gateway. > > Basically I need to check the last 3 known values and based on that raise > alert. > most recent count > 0 : raise alert warning > most recent

[prometheus-users] Grafan report for SLA using blackbox exporter and prometheus

Hello Everyone, I have a requirement to show my system's up time as a report, so for that I am using Blackbox exporter, Prometheus and Grafana. Till now I was showing my system's up time in percentage for specific time range using Grafana with following expression *avg_over_time*(*prob

[prometheus-users] Re: How to define exact hostname or IP in prometheus.yml file

Thanks Brian,I will try the below suggested one and update you. On Sunday, June 14, 2020 at 2:03:11 AM UTC+5:30, Brian Candler wrote: > > See https://www.robustperception.io/controlling-the-instance-label > - you can set the instance label separately from the __address__ that is > scraped > > -

Re: [prometheus-users] Preventing data loss from poor network communication

Alright, I'll look into it. Just in case we don't have the resources required to run prometheus and thanos sidecar on the metrics server. Would there be any issues using the pushgateway to cache the metrics while the network is down? I understand that it would be more complicated to implement, bu

Re: [prometheus-users] Last value(metric count) of a range vector

Yeah it is pushed to pushgateway.. so here for last 4 days no metric was pushed to gateway. Basically I need to check the last 3 known values and based on that raise alert. most recent count > 0 : raise alert warning most recent 2 consecutive count > 0 : raise alert severe most recent 3 consecuti

[prometheus-users] Re: Last value(metric count) of a range vector

Yeah it is pushed to pushgateway.. so here for last 4 days no metric was pushed to gateway. Basically I need to check the last 3 known values and based on that raise alert. most recent count > 0 : raise alert warning most recent 2 consecutive count > 0 : raise alert severe most recent 3 consecuti

Re: [prometheus-users] Issues sending alerts through SendGrid

How can I enable the debug log? I tried looking in the configuration section and found no option for that: prometheus.io/docs/alerting/latest/configuration/ Regarding SendGrid behaviour with incorrect credentials i expect for it to fail, but currently have no got the result. I tried the intera

[prometheus-users] Re: Last value(metric count) of a range vector

that max_over_time or min_over_time can not be applied in my case , because I need recent value.. not the max or min value in that period.. On Monday, 15 June 2020 15:15:03 UTC+5:30, Baitanik Talukder wrote: > > Okay.. if we try to do the workaround with recording rule.. please tell me > how i

Re: [prometheus-users] Last value(metric count) of a range vector

On Mon, 15 Jun 2020 at 09:18, Baitanik Talukder wrote: > I need the last value in a range vector. > Something like > > *last_value*(failure_count{job="xyz"}[5d]) > > it is need for the following scenrio: > > for last 4 days there was no failure_count metric pushed.. > Can you explain more ab

[prometheus-users] Re: Last value(metric count) of a range vector

Okay.. if we try to do the workaround with recording rule.. please tell me how in this case.. just to get the latest know values from this blog ( https://roidelapluie.be./blog/2019/02/21/prometheus-last/) only this one is enough .. right ? - record: available_sms_last expr: available_sms or ava

Re: [prometheus-users] Issues sending alerts through SendGrid

Can you enable debug logging for Alertmanager? That should show whether it is even trying to send an email. How does SendGrid behave when you submit something with invalid credentials? Do you expect that to explicitly fail? >From this doc

Re: [prometheus-users] Last value(metric count) of a range vector

On 15 Jun 01:18, Baitanik Talukder wrote: > I need the last value in a range vector. > Something like > > *last_value*(failure_count{job="xyz"}[5d]) > > it is need for the following scenrio: > for last 4 days there was no failure_count metric pushed.. so last value we > have 5 days back. > outp

[prometheus-users] Last value(metric count) of a range vector

I need the last value in a range vector. Something like *last_value*(failure_count{job="xyz"}[5d]) it is need for the following scenrio: for last 4 days there was no failure_count metric pushed.. so last value we have 5 days back. output of failure_count{job="xyz"}[5d] I need the last value (