Monitored system components
IBM® Control Center can monitor the status and details of system components such as the Cassandra database cluster, the ZooKeeper management ensemble, the replication servers that are required for replication, the storage repository, and the WebSphere® MQ messaging middleware.
The system components that are not internal to Global Mailbox are listed as external systems in the
configuration. If external systems event notifications are enabled, IBM Control
Center receives event notifications to
monitor the status of the following Global Mailbox
system components:
- Cassandra cluster
- An event notification is sent to IBM Control
Center if one of the following
situations occurs involving the Cassandra cluster:
- The Global Mailbox node successfully connects to the Cassandra cluster initially or after a failure is reported.
- The Global Mailbox node cannot connect to the Cassandra cluster.
- The Cassandra cluster cannot maintain
EACH_QUORUM
consistency level.Restriction: It may be possible to satisfyLOCAL_QUORUM
on the same node. An event is raised because there is an issue with consistency in general for this node. - The Cassandra cluster can satisfy the most restrictive consistency, which is
EACH_QUORUM
for both the Global Mailbox management node and the GMCA. - A query to Cassandra times out without completing.
- A query to Cassandra completes successfully after a time out is reported.
- ZooKeeper ensemble
- An event notification is sent to IBM Control
Center if one of the following
situations occurs involving the ZooKeeper ensemble:
- The Global Mailbox node successfully connects to the ZooKeeper ensemble initially or after a failure is reported.
- The Global Mailbox node cannot connect to the ZooKeeper ensemble.
- ZooKeeper successfully creates a lock the first time, or after a failure.
- ZooKeeper cannot create a lock because the ensemble cannot maintain quorum.
- Replication servers
- An event notification is sent to IBM Control
Center if one of the following
situations occurs involving a specific replication server:
- The Global Mailbox node successfully connects to a specific replication server initially or after a failure is reported.
- The Global Mailbox node cannot connect to a specific replication server.
- Replication server starts replicating a file to a data center.
- Replication server completes replicating a file to a data center.
- A segment is successfully replicated.
- An attempt to replicate a segment fails.
Remember: A node in particular data center is dependent on replication servers in other data centers and not replication servers in the same data center. This is because payloads are pulled in using remote replication servers instead of the local replication server. A node heartbeat lists the replication servers in all other data centers, not the ones in the local data center. - Storage repository
- An event notification is sent to IBM Control
Center if one of the following
situations occurs involving the shared storage disk. The notification includes the originating data
center and node that was trying to connect to shared storage.
- The Global Mailbox node successfully connects to the storage repository initially or after a failure is reported.
- The Global Mailbox node cannot connect to the storage repository.
- WebSphere MQ
- An event notification is sent to IBM Control
Center if one of the following
situations occurs involving the WebSphere MQ
queue. The notification includes the originating data center and node that was trying to connect to
WebSphere MQ.
- The Global Mailbox node successfully connects to WebSphere MQ initially or after a failure is reported.
- The Global Mailbox node cannot connect to WebSphere MQ.