4. What’s Fixed in 7.2.0

This section describes bugs fixed in Connext 7.2.0.

• 4.1. Discovery
  • 4.1.1. SPDP2 participants may not have completed discovery if IP mobility event occurred during discovery
  • 4.1.2. Crash if initial_peers sequence contained a NULL string
  • 4.1.3. Failure to deserialize participant discovery information incorrectly allowed discovery to complete
  • 4.1.4. Unbounded memory growth when creating/deleting DomainParticipants
• 4.2. Serialization and Deserialization
  • 4.2.1. Unbounded memory growth when deserializing SPDP discovery sample
  • 4.2.2. Wrong error message when deserializing PropertyQos property value and exceeding property_string_max_length resource limit
  • 4.2.3. Potential unexpected behavior or crash when deserializing SPDP discovery sample
• 4.3. Debuggability
  • 4.3.1. Instance State Consistency QoS was commented out when printed out as XML from code
  • 4.3.2. DataWriter instance statistics were not updated in all cases
• 4.4. Transports
  • 4.4.1. Connext started before Windows completed duplicate address detection on network interfaces
  • 4.4.2. Ungracefully terminated QNX processes using SHMEM transport prevented startup of new processes due to unclosed POSIX semaphores
  • 4.4.3. QNX applications using shared-memory transport may have led to thread priority inversion issues
  • 4.4.4. Stalled communication when using shared-memory transport
  • 4.4.5. Overflow in default TransportMulticastMappingQosPolicy procedure
• 4.5. Reliability Protocol and Wire Representation
  • 4.5.1. Samples lost by reliable reader acknowledging samples it did not receive after remote writer update
  • 4.5.2. Inconsistent RTPS protocol versions broadcasted by Connext
  • 4.5.3. Sample loss when using asynchronous publisher due to missing GAP
• 4.6. Content Filters and Query Conditions
  • 4.6.1. Instance handling on a DataReader and filtering operations in ContentFilteredTopics, QueryCondition, TopicQueries, and Multi-Channel DataWriters may have failed
• 4.7. Dynamic Data
  • 4.7.1. Problems with int8/uint8 support
  • 4.7.2. Connext did not print array dimensions for aliases that were arrays
• 4.8. Performance and Scalability
  • 4.8.1. Performance degradation when using FlatData with ContentFilteredTopics
  • 4.8.2. Performance issues when using FlatData with payload encryption or compression
  • 4.8.3. Transport utilization metrics overflowed in applications with high throughput
• 4.9. APIs (C or Traditional C++)
  • 4.9.1. Some DDS_TypeCode operations may have crashed when invalid arguments were used
  • 4.9.2. Several C API DDS_GUID functions did not account for NULL parameters correctly
• 4.10. APIs (Modern C++ API)
  • 4.10.1. Unexpected rti.connextdds.PreconditionNotMetError when setting optional string members in QoS policies
  • 4.10.2. Move constructors for some of the built-in topic-types were incorrectly implemented
  • 4.10.3. Manually closing some built-in readers could lead to a crash
  • 4.10.4. Incorrect implementation of DynamicDataMemberInfo constructor and assignment may have led to undefined behavior
  • 4.10.5. int8_t, uint64_t, int64_t not supported as primitive types in C++11 (Modern C++) Dynamic Type API
  • 4.10.6. Policy getter for rti::core::policy::Monitoring previously missing
• 4.11. APIs (Java)
  • 4.11.1. Possible memory leak in DynamicData copy constructor
  • 4.11.2. Some ReliabilityQos methods did not consider the instance state consistency QoS
• 4.12. APIs (Python)
  • 4.12.1. Access to collection elements in some DynamicData accessors was not zero-based
• 4.13. APIs (Multiple Languages)
  • 4.13.1. Looking up a DataReader using the wrong class in Modern C++ or Python did not raise clear exception
  • 4.13.2. Alias type not obtainable using a QosProvider
• 4.14. XML Configuration
  • 4.14.1. Incorrect parsing of data_representation attribute in XML type definitions
  • 4.14.2. Creating Topic-specific entities from a <qos_profile> using QoS profile inheritance and/or composition returned incorrect values
  • 4.14.3. configuration_variables tag was not effective
  • 4.14.4. Using languageBinding attribute on union types in XML caused parsing error
• 4.15. Instances
  • 4.15.1. Instances transitioned due to instance state consistency did not respect propagate_dispose_of_unregistered_instances
  • 4.15.2. Instance purging based on source timestamp did not work
• 4.16. Crashes
  • 4.16.1. Race condition when using multiple threads to enable same DomainParticipant
  • 4.16.2. Possible crash gathering periodic metrics for a resource that was being added or deleted at the same time
  • 4.16.3. Potential crash when configuring logging verbosity to NDDS_CONFIG_LOG_VERBOSITY_STATUS_LOCAL or higher
  • 4.16.4. Malloc called when handling SIGSEGV
  • 4.16.5. Calling delete_contained_entities APIs could cause a crash in the thread that collects periodic metrics
  • 4.16.6. Application could crash when disabling and re-enabling Monitoring Library 2.0 due to internal error
  • 4.16.7. Low-memory conditions could lead to crash on several platforms if allocation of high resolution clock failed
• 4.17. Entities
  • 4.17.1. Monitoring Library 2.0 incorrectly collected both enabled and disabled DDS Entities
  • 4.17.2. Application may have hung when deleting a monitored DDS entity
  • 4.17.3. Monitoring Library 2.0 did not assert disabled DDS Entities when the Entities were enabled
• 4.18. Interoperability
  • 4.18.1. Possible incomplete endpoint discovery when communicating with other DDS vendors
• 4.19. Vulnerabilities
  • 4.19.1. Out-of-bounds read while deserializing malformed partition parameters from malicious RTPS message
  • 4.19.2. Out-of-bounds read while deserializing malformed IPv6 locator from malicious RTPS message
  • 4.19.3. Remote modification of DomainParticipant names in unsecure system
• 4.20. Other
  • 4.20.1. Possible hang in application if something failed while adding a new observable resource
  • 4.20.2. Native Android applications were not shipped
  • 4.20.3. Error creating a DataWriter using durable writer history if setting property dds.data_writer.history.odbc_plugin.builtin.sample_cache_max_size to -1
  • 4.20.4. References to missing header file in Connext Professional source bundle
  • 4.20.5. Application may have hung when event and event snapshot were published simultaneously for same observable resource
  • 4.20.6. Access to an internal field of observable resources was not thread safe
  • 4.20.7. Deadlock issue resolved when disabling Monitoring Library 2.0 during command processing
  • 4.20.8. Running rtisetenv_<arch>.bat caused issues in PATH environment
  • 4.20.9. Unable to start Launcher, Admin Console, Code Generator, and Monitor in Windows when the RTI Workspace contained white spaces