I want to use this command to backup my entire influxdb2 database on a weekly basis:
influx backup <backup-path> -t <root-token>
This generate a bunch of files in a directory (currently totaling about 3GB), however I want to then move these across to an external drive. Do the files from each run of the backup command get completly replaced at each run or is it incremental in anyway ? I want to avoid copying large amounts of data to the external drive each time if I can avoid it.
Thanks
I’ve done a bit more digging and it appears the backup is split into “shards” of 7days worth of data. Multiple runs seems to produce reproducible results of outputs files that match in terms of the hash, however they are named according to the day the backup was run. This means, it will generate a complete “new” set of files after each run.
I could run a post-processing script to rename the files and update the manifest, however it seems very error prone.
Is there a way to fix the file naming in someway not tied to the date they were created ? or recommend a backup strategy ?
Thanks
Is there a better place to get help than this forum ?
