How can I use `rest/api/content` to download all the 800k pages of my Confluence wiki without timing

[Franck Dernoncourt]

I  want to download all the 800k pages of my Confluence wiki.

I'd like to use:

curl -u wikiusername:wikipassword https://wiki.hostname.com/rest/api/content?start=1`

and simply increase start from 1 to 800000.

However, the response time increases as start increases, and from ~80000 begins to timeout:

start	response time (seconds)
1	0.4
1,000	2.5
10,000	9
50,000	112
100,000	timeout

How can I use rest/api/content to download all the 800k pages of my Confluence wiki without timing out?

[Nic Brough -Adaptavist-]

This is because when you make the REST call, the server is trying to build the response in memory before it can send it back.  If you make it too large, the process will fail.

You're going to need to page through what you're trying to download, you can't do it in one massive great chunk (unless you increased the server memory to something massive)

I'd also want to quickly question why?  What is this download going to do for you?  I'm thinking there may be a better option (like parsing a backup)