Hey everyone, I’m looking into doing a full reindex of our dataset every day or so. I saw this link in the docs mentioning that I should use a Collection Alias to “do a reindex without any downtime”, then switch the alias to the new collection. Is there anyone that can elaborate on how reindexing without using the alias feature could cause downtime and in what way? Thanks :-))

Hmm, that downtime comment in the docs is actually a bit out of place in the context of the data syncing guide you're talking about. Usually you'd use an alias when you want to change the schema in a zero downtime way. If you're just reindexing documents using an upsert, then just calling the bulk import endpoint with action=upsert won't incur downtime

@Jason Bosco Thanks for clarification, much appreciated 🙏

I've already updated the docs, let me know if this helps avoid that confusion now: