Navigation
This version of the documentation is archived and no longer supported.

Change the Size of the Oplog

New in version 3.6.

This procedure changes the size of the oplog on each member of a replica set using the replSetResizeOplog command, starting with the secondary members before proceeding to the primary.

Important

You can only run replSetResizeOplog on replica set member’s running with the Wired Tiger storage engine.

Perform these steps on each secondary replica set member first. Once you have changed the oplog size for all secondary members, perform these steps on the primary.

A. Connect to the replica set member

Connect to the replica set member using the mongo shell:

mongo --host <hostname>:<port>

Note

If the replica set enforces authentication, you must authenticate as a user with privileges to modify the local database, such as the clusterManager or clusterAdmin role.

B. (Optional) Verify the current size of the oplog

To view the current size of the oplog, switch to the local database and run db.collection.stats() against the oplog.rs collection. stats() displays the oplog size as maxSize.

use local
db.oplog.rs.stats().maxSize

The maxSize field displays the collection size in bytes.

C. Change the oplog size of the replica set member

To change the size, run the replSetResizeOplog passing the desired size in megabytes as the size parameter. The specified size must be greater than 990, or 990 megabytes.

The following operation changes the oplog size of the replica set member to 16 gigabytes, or 16000 megabytes.

db.adminCommand({replSetResizeOplog: 1, size: 16000})

D. (Optional) Compact oplog.rs to reclaim disk space

Reducing the size of the oplog does not automatically reclaim the disk space allocated to the original oplog size. You must run compact against the oplog.rs collection in the local database to reclaim disk space. There are no benefits to running compact on the oplog.rs collection after increasing the oplog size.

Important

The replica set member cannot replicate oplog entries while the compact operation is ongoing. While compact runs, the member may fall so far behind the primary that it cannot resume replication. The likelihood of a member becoming “stale” during the compact procedure increases with cluster write throughput, and may be further exacerbated by the reduced oplog size.

Consider scheduling a maintenance window during which writes are throttled or stopped to mitigate the risk of the member becoming “stale” and requiring a full resync.

Do not run compact against the primary replica set member. Connect a mongo shell to the primary and run rs.stepDown(). If successful, the primary steps down and closes all open connections. Reconnect the mongo shell to the member and run the compact command on the member.

The following operation runs the compact command against the oplog.rs collection:

use local
db.runCommand({ "compact" : "oplog.rs" })

If the disk space allocated to the original oplog size is not reclaimed, restart mongod and run the commands from step D again. Restarting mongod runs recalculations in WiredTiger that might allow compact to release more space to the OS.

For clusters enforcing authentication, authenticate as a user with the compact privilege action on the local database and the oplog.rs collection. For complete documentation on compact authentication requirements, see compact Required Privileges.