Log rotation

logrotate is a tool to manage the log file growth. MongoDB logs can grow quickly. Without rotation, logs may fill the filesystem and disrupt the database. logrotate ensures predictable log retention and disk usage.

logrotate runs in a logcollector sidecar container within each database Pod. It reads the log files at the specified path and rotates them according to the set of rules.

By default, logrotate in Percona Operator for MongoDB works as follows:

• Rotates the /data/db/logs/mongod.full.log daily.
• If the log file exceeds 100 MB, it will be rotated on the next run, regardless of the schedule.
• Keeps up to 7 rotated log files.
• Skips missing or empty log files.
• Leaves rotated logs uncompressed.
• Copies the current file to the rotated file and then truncates this file to zero in place.
• Ensures pre-rotation scripts run only once per rotation with sharedscripts.

• Before rotating, performs the following in the prerotate script:

• Runs the db.adminCommand({ logRotate: 1 }) command so that MongoDB closes its current log and starts a new one.

• Deletes any mongod.log.* files in /data/db/logs/ that are older than 7 days.

Configure log rotation

You can customize log rotation if you need different retention, size limits, or want to rotate additional files (for example, audit logs). This gives you predictable disk usage and keeps compliance or troubleshooting logs available for as long as you need.

You can configure log rotation in the following ways:

• Override the default logrotate configuration via the Custom Resource
• Add additional configuration via ConfigMap
• Set a custom log rotation schedule

Regardless of the method you use, after you apply the new configuration, the Operator will restart the database Pods.

Override the default logrotate configuration

Use the logcollector.logrotate.configuration section in the Custom Resource to completely override the default logrotate settings. For example, you can configure it to rotate logs when their size is between 100 KB (minimum) and 200 MB (maximum).

Important

You must provide the full logrotate configuration because the Operator replaces the default configuration with the one you provide. Refer to the default configuration to see the built-in logrotate rules and use it as a guide for your custom settings.

Here’s an example configuration:

spec:
  logcollector:
    logrotate:
      configuration: |
        /data/db/logs/mongod.full.log {
           daily
           minsize 100K
           maxsize 200M
           rotate 7
           missingok
           nocompress
           notifempty
           copytruncate
           sharedscripts
           prerotate
               # rotate mongod.log using 'db.adminCommand({ logRotate: 1 })'
               mongosh "mongodb://${MONGODB_USER}:${MONGODB_PASSWORD}@${MONGODB_HOST}:${MONGODB_PORT}/admin" \
                       --eval 'db.adminCommand({ logRotate: 1 })'
               find /data/db/logs/ -type f -name 'mongod.log.*' -mtime +7 -delete
           endscript
        }

Apply the configuration:

kubectl apply -f deploy/cr.yaml -n <namespace>

Add extra logrotate configuration

You can pass additional configuration options for the default logrotate configuration via a ConfigMap or a Secret. Use logcollector.logrotate.extraConfig.name to load an additional .conf file from a ConfigMap or Secret. The file name must end with .conf.

Important

The mongodb.conf name is reserved for the main configuration and you must not use it in your custom ConfigMap or Secret.

For example, you want to also rotate audit log files. Here’s how to provide this additional configuration.

1. Create a ConfigMap configuration file. For example, auditlog.yaml.

   apiVersion: v1
   kind: ConfigMap
   metadata:
     name: logrotate-custom
   data:
     audit.conf: |
       /data/db/audit/*.json {
          minsize 10K
          maxsize 20K
          rotate 10
          missingok
          nocompress
          notifempty
          sharedscripts
          copytruncate
         }
   

2. Create the ConfigMap:

   kubectl apply -f auditlog.yaml
   

3. Reference the ConfigMap in your Custom Resource and specify the path to audit logs:

   spec:
     # ...
     replsets:
     - name: rs0
       size: 1
       configuration: |
         auditLog:
           destination: file
           format: JSON
           path: /data/db/audit/audit.json
     logcollector:
       logrotate:
         extraConfig:
           name: logrotate-custom
   

4. Apply the configuration:

   kubectl apply -f deploy/cr.yaml -n <namespace>
   

Set a custom schedule

Starting with version 1.22.0, the default logrotate schedule changed from running once per hour to running once per day.

Use the logcollector.logrotate.schedule option in the Custom Resource to set a custom rotation schedule as a cron expression. For example, set the new rotation time to 2:15 a.m every Sunday:

spec:
  logcollector:
    logrotate:
      schedule: "0 15 2 * * 0"

Apply the configuration:

kubectl apply -f deploy/cr.yaml -n <namespace>

---

Last update: February 20, 2026
Created: February 20, 2026