[MyLockServer] Sections

The [MyLockServer] configuration sections contain details of your Redis servers to use for flushing and locking.

Flushing is the process of writing cached data to disk during indexing. It occurs periodically, and it is resource-intensive. It is important to prevent two flushes from occurring simultaneously.

If a lock file exists, a server writes information to it to notify other servers that it is flushing, and other servers wait until it is finished before writing to the lock file and performing their flushes.

You can configure a lock file by using the FlushLockFile configuration parameter in the [Server] section. However, HPE does not recommend that you use a FlushLockFile if you are using networked storage.

For networked storage, you can configure the [FlushLock] and [MyLockServer] sections with the details of Redis servers. In this case, the Content component attempts to contact all the configured Redis servers to obtain a lock, and flushes only if it can obtain a lock from more than half of the Redis servers.

For failover purposes, HPE recommends that you configure at least three Redis servers for locking. In this case, if one of the servers is unavailable, the Content component can still obtain more than half of the locks.


Your lock servers must use Redis version 2.8 or later.

1=LockServer2 2=LockServer3


Port=6389 [LockServer3] Host= Port=6389