How do create a threshold for read_only_allow_delete?

Hello! I am running to an issue where my indices lock after it reaches a certain limit. Where can I find the limit and/or how can I adjust it so that it only locks when it reaches 5% of disk space used?

You are looking for the flood-stage watermark, documented on the page on disk-based shard allocation. It defaults to 95%.

Do you mean "free" instead of "used"? Marking indices as readonly at 5% disk usage seems like a strange choice. The default is to mark them readonly at 95% disk usage (i.e. 5% disk free).

Yes correct I meant "free" space. I am currently using the following command to unlock it.
PUT _settings
"index": {
"blocks": {
"read_only_allow_delete": "false"

How do I update this to a 3% usage lock in the elastic search .yml file?

Add this line, as per the docs I shared above:

cluster.routing.allocation.disk.watermark.flood_stage: 97%

1 Like

Ah, I see, you can use any of these lines in the yml file :slight_smile:

Very good to know, thanks David!

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.