Audit records are waiting
Weik.io 2026.2A backend node has audit records that have waited a minute to reach the audit log. After 15 minutes the problem is an error. It's an error at once when the node can't keep new records on its disk.
| Kind | Severity | Who sees it |
|---|---|---|
audit- |
Warning after a minute, and an error after 15 minutes | Platform administrators, in a list that isn’t narrowed to an environment or a project |
Each backend node with waiting records has its own row in the problem list, such as Audit records waiting on the backend node on hub server 2. The row has a figure such as 12 audit records waiting, or 12 audit records waiting, 3 in the log only when some new records went only to the node’s log.
What it means#
A backend node has audit records that aren’t in the audit log yet. The node keeps them and sends them when it can. Until then, the audit log doesn’t show those changes.
What raises it#
The oldest waiting record has waited a minute. It becomes an error after 15 minutes.
The problem shows at once as an error while a waiting record is only in the node’s memory, because the node’s data volume didn’t take it. It also shows at once as an error while the node keeps no new records. The node stops keeping them when it already keeps 100,000 records or 256 MiB, or when the disk under its data folder is down to a tenth free.
What to check#
See every part of the hub:
$ weikio platform healthRead the node’s warnings:
$ weikio platform logs backend --since 15m --level warnHow to fix it#
Check that the hub’s data servers are running and can reach each other. When a record is only in the node’s memory, check the backend’s data volume. When you fix the cause, the node sends the records that waited.