Hi, is there an option to aggregate SigNoz data fr...
# general
a
Hi, is there an option to aggregate SigNoz data from multiple SigNoz instances into a single one? I have multiple k8s clusters, I want to have a SigNoz deployment on each of them (to monitor a single cluster they been deployed at), and in addition I want to have a centralized SigNoz instance that will aggregate all the โ€˜singleโ€™-SigNoz data. I also want to be able to navigate between the different SigNoz instances-data. Today Im implementing this setup using prometheus & thanos & grafana. Thanks ๐Ÿ™‚
a
This should be possible with OpenTelemetry Collectors. You can run OpenTelemetry Collectors in each of your k8s clusters, and then send the data collected by each OpenTelemetry Collector to a single SigNoz instance. Regarding navigating between different SigNoz instance s data, that should be possible with environment variables. @Prashant Shahi might be able to shed more light on this.
p
Hey @Asaf Fadida ๐Ÿ‘‹ We have
k8s-infra
chart just for that. You can have
signoz
chart installed in one K8s cluster, while
k8s-infra
chart in other K8s clusters.
k8s-infra
chart collects metrics and logs in any external cluster and sends to SigNoz.
a
Thats great, thanks! What if I want to have multiple SigNoz instances? One for every k8s cluster (SigNoz will be installed on the monitored k8s cluster) + additional SigNoz instance that gets all the other k8s clusters data. Is it possible?
@Prashant Shahi
p
It is possible, but that would be independent signoz clusters.
that would be no related to each other.
there would be many signoz instances. If you desire that, you can go ahead with that as well.
But it's not recommended to have N signoz instances for N K8s clusters.
a
Works for me. The question is if otel-collector can send metrics to multiple targets (one local-Signoz + one remote SigNoz)
p
yes, it can.
a
Perfect. Thanks
p
But would need to make changes in the config manually.
a
๐Ÿ™
a
Asaf, can you elaborate more on your use case? Why do you want to run local SigNoz instances for each cluster?
a
Our system might be deployed on both airgapped-on-prem & non-airgapped environments. We want to provide a monitoring solution within the product-deployment itself (deploying SigNoz on the monitored cluster), and where we can (non-airgapped), we would like to see the monitoring-related-data on a single place (centralized SigNoz)
Make sense?
a
Got it. Makes sense, Asaf.