Known Issues
Global Mailbox v6.2.0 has the following known issues:
| Functionality | Known issue | Workaround |
|---|---|---|
| Installation | Starting all Global Mailbox servers simultaneously during initial setup can lead to replication failures. | Global Mailbox servers must be started one at a time during initial setup. |
| Initial setup | There are port conflict errors or security concerns observed for the port number 9160. It is the port number used by Thrift RPC server in Cassandra. |
Cassandra doesn’t use the Thrift RPC server any longer. If you face security concerns or any other issue on the port number 9160, you can edit the cassandra.yaml from the location <cassandra install location>/conf and set start_rtc:false. |
| Initial setup | When trying to start admin servers at the same time during the initial setup, a 1970 date is written into the replication_queue_ptr table in the database and files are not replicated. | Start Global Mailbox Admin servers one at a time during the initial setup (while the replication_read_ptr table is empty or after it has been truncated). After the initial setup, you can start Global Mailbox Admin servers at the same time. |
| Administrator user interface | The administrator UI for Global Mailbox cannot be reached from the Sterling B2B Integrator administrator UI when all replication servers in a data center are manually suspended. Error dialogs indicate the Global Mailbox client adapter cannot be found. | Administrative activity for Global Mailbox must be performed by directly accessing the Global Mailbox administrative UI. |
| ASCP | ASCP frequently generates logs about session shutdown failures. | Ignore Session shutdown failure errors. This is a normal condition. |
| Standby queue manager | The Event Rule Adapter configuration shows an empty value for the Connection Name list rather than listing the active and standby queue managers. | Configure the Global Mailbox Event Rule Adapters in Sterling B2B Integrator to use a connection list with both the active and standby queue managers in the list. You should only list queue managers in the same datacenter as the Sterling B2B Integrator node that you are configuring. |
| Performance | Due to limitations in the Cassandra Repair jobs, Global Mailbox can process 500,000 messages per day. | If your production environment must support a higher load of messages per day, contact IBM Support. |
| Performance | The highest measured peak throughput with Global Mailbox is 70 messages per second. | If your production environment must support a higher load of messages per day, contact IBM Support. |
| Performance | Each Global Mailbox server consumes 1 CPU core while idle. | Allocate sufficient CPU resources to all servers. |
| Non-English topics | Information about firewall configuration was not translated. | Ensure that your firewalls are configured to allow connections to the ports for ZooKeeper, Cassandra, Global Mailbox Management, WebSphere® MQ, and Sterling B2B Integrator nodes. |
| Non-English topics | Updated information about recovering from a planned or unplanned data center outage was not translated. | Use the English version of the topics Restarting Cassandra nodes after data center maintenance and Restarting the Global Mailbox management node. |
| Non-English topics | Updated information for configuring advanced event processing was not translated. | Use the English version of the topic Configuring WebSphere MQ advanced event management. |