I have an existing InfluxDB server that I’m wanting to deploy influxdb-relay in front of. What’s the best process for initially replicating data from my existing server to two additional new servers I want to run beside it so I have a total of 3 servers sitting behind my 2 influxdb-relay servers?
I’m assuming I can use the InfluxDB backup and restore procedures to replicate data to the new servers, but do I need to keep writes from happening on the existing server while I do that and get influxdb-relay up and running, or can the two new servers be a bit out-of-sync with the existing server when the influxdb-relay deployment starts being used by clients?
@activeshadow What I would do is setup the relay with 2 new servers behind it then switch the DNS or routing to direct clients to the new setup and use influx_inspect and influx -import to write the data to the new servers.
Ah… OK. Let me restate what I think you’re saying…
Let the new servers be written to while backfilling them with data from the existing server, then just remove the existing server.
Correct?
And just to confirm, did you mean to say influx_inspect export (as opposed to just influx_inspect)?
@activeshadow Yup! I’ve done this migration strategy on a couple of internal instances we have and it works great.
Awesome, thanks @jackzampolin. I’ll give it a go early next week and report back here as to how it went and any issues I encountered.
1 Like
