Bug ID 1644545: Fluentd repeatedly runs out of memory due to missing elasticsearch index

Last Modified: Oct 19, 2025

Affected Product(s):
BIG_IP_NEXT(CM) MA-CM(all modules)

Known Affected Versions:
20.3.0

Opened: Sep 10, 2024

Severity: 1-Blocking

Symptoms

- Fluentd OOM messages are seen on the BIG-IP Next Central Manager console and in the kernel logs - You are unable to log into BIG-IP Next Central Manager via the UI or API - BIG-IP Next Central Manager backup fails - BIG-IP Next Central Manager qkview does not produce a qkview

Impact

BIG-IP Next Central Manager application logs are not persisted in elasticsearch and BIG-IP Next Central Manager operation is compromised.

Conditions

Perform a fresh install of BIG-IP Next Central Manager using the 20.3.0 release.

Workaround

1. SSH into BIG-IP Next Central Manager. 2. Run the command: kubectl exec cmdb-elasticsearch-0 -- curl -sX PUT "http://localhost:9200/application_cm" -H 'Content-Type: application/json' -d '{ "settings": { "number_of_shards": 1, "number_of_replicas": 0 }, "mappings": { "properties": { "@timestamp": { "type": "date" }, "_hash": { "type": "keyword" }, "action": { "type": "keyword" }, "level": { "type": "keyword" }, "msg": { "type": "text" }, "podname": { "type": "keyword" }, "source": { "type": "keyword" } } } }' 3. Run the command: kubectl delete pod fluentd-0 4. Run the command: kubectl wait --for=condition=Ready pod/fluentd-0 --timeout 5m 5. If authentication via the UI is failing, reload the login page.

Fix Information

None

Behavior Change

Guides & references

K10134038: F5 Bug Tracker Filter Names and Tips