shift-rmm VS zincobserve

I have just published a dashboard to the grafana marketpalce that includes showing failed services. this comes with shift-mon

I love Victoriametrics and have an ansible role that setup Victoriametrics, Grafana, and Loki using podman on a Linux VM, and is super easy to run. The only thing is switching to datadog to on prem is your team might not have the time available to manage the system or setup something to watch the watcher like having the VM server send metrics to Grafana cloud.

Grafana cloud has a two week trial that should give you a good indication of what your usage is. If you have spare hardware self hosting can be easy here is my project that turns a vm into a monitoring stack based around grafana

This is query I use for host down alerts its not perfect source

Proxmox has the ability to ship metrics to influxdb or graphite without installing an agent. I use shift-mon which uses Victoriametrics and grafana under the hood

I use and love Victoriametrics, Telegraf, and grafana for monitoring proxmox, docker, and other infrastructure. Here is a repo that has an Ansible role to set all of it for you. Grafana cloud also has a free tier that is good for a lot of homelabs

I uploaded my grafana dashboard and extra config for opnsense to grafana's dashboard site. This a uses Victoriametrics for metrics, Loki for logs, and grafana for visualization. The Ansible roles to set it all up are found on gitlab

2 and 3 You can use the GUI in grafana to build Prometheus queries in grafana. I also prefer push over pull. I found out Victoriametrics is a thing and let's you push data as if it was influxdbv1 and add it to grafana as a promtheus data source. I deploy everything via ansible and here is the link to the repo

Prometheus might not be the best tool I prefer Telegraf and Victoriametrics so I can push metrics and still query it as if it where Prometheus. My setup can be found here

Take a look at https://github.com/zinclabs/zincobserve or Zinc cloud It has logs, metrics, traces, dashboards, alerts, functions and more.

For some reason if you do not want to have the complexity of log forwarders you could simply send the json logs using http them to log files in /var/lib/docker/containers . You would then configure vector/fluentbit to send it to appropriate centralized log search tool. A good light weight tool for self hosting would be - https://github.com/zinclabs/zincobserve . From what you have mentioned I think you would rather prefer the cloud service which you can access at https://zinc.dev/

shell wget https://github.com/zinclabs/zincobserve/releases/download/v0.4.4/zincobserve-v0.4.4-linux-amd64-musl.tar.gz tar -zxvf zincobserve-v0.4.4-linux-amd64-musl.tar.gz ./zincobserve

ZincObserve for logs, metrics, traces and dashboards - https://github.com/zinclabs/zincobserve

We are building long term storage for prometheus. Actually we are building an unified system for logs, metrics and traces - see https://github.com/zinclabs/zincobserve . We are working on implementing promql query interface for ZincObserve. We would love to get community feedback on which functions should we prioritize first - https://github.com/zinclabs/zincobserve/issues/582