On-Premises Release Notes
For an upgrade to the new version please follow the Packaged Based Operations manual.
Note: We changed the required users and groups:
- - instana-cashier:instana-cashier
+ - instana-cashier-ingest:instana-cashier-ingest
+ - instana-cashier-rollup:instana-cashier-rollup
Cumulative Features from Release #155, #156, #157
- Host sensor: Add inode metrics for supported filesystems on the host dashboard. Also add a health rule that triggers if inode usage is more than 80 percent.
Instana now provides first-class GraphQL queries, mutations and subscriptions support for Node.js! For more information, see GraphQL.
Sybase SQL Anywhere sensor
Instana now discovers and monitors Sybase (SAP) SQL Anywhere, embedded database for application software. Read more about our Sybase SQL Anywhere support.
- Python sensor: New mysqlclient package instrumentation & visibility
- Python sensor: New PyMySQL package instrumentation & visibility
- Etcd Sensor: Improve metric negative values handling
- MySQL Sensor: MySQL 8 support
- RabbitMQ Sensor: RabbitMQ 3.7.14 support
- HAProxy sensor: Improve stability when encountering a large number of metrics
- WebLogic sensor: Improve stability when encountering a large number of metrics
- Prometheus sensor: Improve stability when encountering a large number of metrics
- Jenkins sensor: Improve stability when encountering a large number of metrics
- Nginx sensor: Performance improvements regarding memory usage
- Python sensor: Update Flask instrumentation (now with rendering visibility and optional blinker support)
- Python sensor: Better performance through snapshot caching
- Python sensor: New module support - try it out with
python -m instana
- Python sensor: Improve thread management & safety
- Python sensor: Better fork detection & handling
- Python sensor: More intelligent package loading semantics
- Ruby sensor: Custom Tracing: Update and improve
- Etcd sensor: Improve the connection URL handling in Docker environment
- Openshift: Display Deployment Configs on more Kubernetes dashboards
- MS SQL sensor: Allow custom rules for the number of lock requests and the number of deadlocks
- Azure sensors: More intelligent handling of Azure REST API rate limits. Multiple service principals can be used in order to handle Azure API rate limits for larger subscriptions. Resource usage optimizations.
- Ceph sensor: Fix issue regarding quorum not reached health rule
- Springboot sensor: Improve gathering of http statuses metrics
- Php Sensor: improved CPU usage.
- Php Tracer: new and improved MongoDb instrumentation.
- Php Tracer: capture errors on IBM DB2 queries.
- Redis Sensor: added memory fragmentation ratio.
- Service Mapping: improved service naming based on JDBC connection string by using the database schema name as the service name. This enhancement provides better support and more consistency across different database technologies.
- When Amazon EC2
Name tags are used to name a group of hosts, the unique host naming fallbacks to FQDN, and then to the hostname. This is visible on the infrastructure map. For more information about how to collect EC2 tags, see the EC2 tags section.
- Agent: sensitive information is redacted from the Docker Container Info. Sensitive information is configured in
configuration.yaml under the
- Improved Trace Detail to precisely show the difference in erroneous calls and error logs.
- Error or warning logs are now displayed in the parent call.
- Settings area: on the right-hand side of the Dynamic-Focus-Query (DFQ) input field, click the Star button to display a list of stored filters. Click each filter to paste it into the input field.
- Icons on the dashboard breadcrumbs, and the infrastructure map sidebar, now reflect the health status of an entity.
- Dynamic Focus: The
entity.selfType:host queries match only the events corresponding to the particular entity type, excluding the events of Application, Service and Endpoint entities.
- Kafka: Fix Partitions count in the
Topics/Partitions for Kafka Cluster
- Updated Kubernetes supported versions
- Alert settings: Fix minor issues when configuring an alert. The state of the selected Events or EventTypes got lost when switching to the other option.
- Php Tracer: show HTTP methods in
- Php Tracer: fixes potential segfault in ElasticSearch instrumentation.
- Php Tracer: INSTANASERVICENAME is now picked up on Apache.
- The deprecated Rules and RuleBindings API to configure Custom Events is no longer supported. Please use /api/events/settings/event-specifications/custom of the Event-Settings API instead.