Ceramic Node Observability

We could share more about best practice of running nodes at different scale here.


Always open to some more discussion on running nodes. I have been standing up a bunch lately so certainly helping to manage plenty.

Hi. Thanks for the reply. Any tips on how we could improve IPFS query performance? Was getting lots of timeouts…

In terms of metrics, do we have existing metrics we could leverage on? eg. traces that shows the breakdown of query latency? (eg. how much does ipfs took in the critical path of the request-response)

@spencer @msena Any metrics or tips we could learn from? From our current deployment, IPFS seems to be a major bottleneck?
Would it be possible to use IPFS cluster or elastic IPFS ? or kubo is the only compatible verison ?