Release 10.5.1
Connectivity and Adapters
PAM-29458
Multi-threaded performance scaling of JMS XPath mapping is very poor.
The previous bottleneck affecting JMS scalability on multicore platforms has now been addressed. JMS now makes better use of cores and delivers greater performance.
Cumulocity IoT Transport
PAM-30999
Runtime error when activating model in Analytics Builder which should be more robust in handling “atypical” objects.A runtime error was seen in Analytics Builder when a model failed to activate in one of the subtenants.
This issue is now resolved.
PAM-31014
Cumulocity chain should log a WARN message if measurement timestamp is not set or invalid.
It is possible to forget setting a timestamp when sending measurements, alarms or events. If the timestamp is unset or is during the year 1970, a warning message is logged once per managed object.
PAM-31036
Cumulocity IoT transport cannot use credentials from environment for long polling.
Previously, the old Cumulocity IoT transport required a user name and password to be specified in a configuration file. Now, it can pick up configurations from the environment. This is especially useful in the microservice mode where the user does not need to populate the configuration file and an empty configuration file can be injected. In the case of an app-key, a lookup for the application using its name is performed. If the lookup fails, then the app-key must be provided in the configuration file.
ADBC
PAM-29763
IAF allows commands to be executed before fully up in ADBC.
To conform with best practice, the ADBCHelper has been updated to ensure the adapter is up before it can proceed with any next operations, in both startup and reconnection scenarios.
Correlator
PAM-27043
Correlator and IAF should print logfile to stdout on startup.
If the correlator or the IAF is configured to log to a file, then the log file being used will be output to stderr. In the case of the correlator, this is the full path and filename whereas the IAF will only log the filename.
Other
PAM-30176
VisualizingDataInADashboard EPL generates huge number of DataView updates.
The Persisting Data and Visualizing Data samples would use up an ever increasing amount of memory and if left running for a long period of time, the process would shut down. This issue is now resolved.