From 14ddf2f2def55fe5f713112a29ccb3f6efff68db Mon Sep 17 00:00:00 2001 From: Luke Whiting Date: Fri, 27 Sep 2024 13:32:23 +0100 Subject: [PATCH] Note in docs about interpreting IO stats when running in docker (#113676) * Note in docs about incorrect IO stats when running in docker * Update docs/reference/cluster/nodes-stats.asciidoc Co-authored-by: David Turner * Requested PR changes to wording * Update docs/reference/cluster/nodes-stats.asciidoc Co-authored-by: David Turner --------- Co-authored-by: David Turner --- docs/reference/cluster/nodes-stats.asciidoc | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/docs/reference/cluster/nodes-stats.asciidoc b/docs/reference/cluster/nodes-stats.asciidoc index 61c58cea95b83..adf8229712ecc 100644 --- a/docs/reference/cluster/nodes-stats.asciidoc +++ b/docs/reference/cluster/nodes-stats.asciidoc @@ -1716,6 +1716,10 @@ See <> for more information about disk watermarks a `io_stats` (Linux only):: (objects) Contains I/O statistics for the node. + +NOTE: These statistics are derived from the `/proc/diskstats` kernel interface. +This interface accounts for IO performed by all processes on the system, even +if you are running {es} within a container. + .Properties of `io_stats` [%collapsible%open]