v0.94.0

This Splunk OpenTelemetry Collector release includes changes from the opentelemetry-collector v0.94.1 and the opentelemetry-collector-contrib v0.94.0 releases where appropriate.

🛑 Breaking changes 🛑

• (Splunk) The Splunk OpenTelemetry Collector Windows install script
  now installs the Splunk Distribution of OpenTelemetry .NET
  instead of the SignalFx Instrumentation for .NET
  when the parameter -with_dotnet_instrumentation is set to $true (#4343)
• (Core) receiver/otlp: Update gRPC code from codes.InvalidArgument to codes.Internal when a permanent error doesn't contain a gRPC status (#9415)
• (Contrib) kafkareceiver: standardizes the default topic name for metrics and logs receivers to the same topic name as the metrics and logs exporters of the kafkaexporter (#27292)
  If you are using the Kafka receiver in a logs and/or a metrics pipeline
  and you are not customizing the name of the topic to read from with the topic property,
  the receiver will now read from otlp_logs or otlp_metrics topic instead of otlp_spans topic.
  To maintain previous behavior, set the topic property to otlp_spans.
• (Contrib) pkg/stanza: Entries are no longer logged during error conditions. (#26670)
  This change is being made to ensure sensitive information contained in logs are never logged inadvertently.
  This change is a breaking change because it may change user expectations. However, it should require
  no action on the part of the user unless they are relying on logs from a few specific error cases.
• (Contrib) pkg/stanza: Invert recombine operator's 'overwrite_with' default value. (#30783)
  Previously, the default value was oldest, meaning that the recombine operator should emit the
  first entry from each batch (with the recombined field). However, the actual behavior was inverted.
  This fixes the bug but also inverts the default setting so as to effectively cancel out the bug fix
  for users who were not using this setting. For users who were explicitly setting overwrite_with,
  this corrects the intended behavior.

🚩 Deprecations 🚩

• (Core) configgrpc: Deprecate GRPCClientSettings, use ClientConfig instead (#6767)

💡 Enhancements 💡

• (Splunk) Add a resource attribute to internal metrics to track discovery usage (#4323)
• (Splunk) Create a multi-architecture Windows docker image for the collector (#4296)
• (Splunk) Bump splunk-otel-javaagent to v1.30.2 (#4300)
• (Core) mdatagen: Add a generated test that checks the config struct using componenttest.CheckConfigStruct (#9438)
• (Core) component: Add component.UseLocalHostAsDefaultHost feature gate that changes default endpoints from 0.0.0.0 to localhost (#8510)
  The only component in this repository affected by this is the OTLP receiver.
• (Core) confighttp: Add support of Host header (#9395)
• (Core) mdatagen: Remove use of ReportFatalError in generated tests (#9439)
• (Contrib) receiver/journald: add a new config option "all" that turns on full output from journalctl, including lines that are too long. (#30920)
• (Contrib) pkg/stanza: Add support in a header configuration for json array parser. (#30321)
• (Contrib) awss3exporter: Add the ability to export trace/log/metrics in OTLP ProtoBuf format. (#30682)
• (Contrib) dockerobserver: Upgrading Docker API version default from 1.22 to 1.24 (#30900)
• (Contrib) filterprocessor: move metrics from OpenCensus to OpenTelemetry (#30736)
• (Contrib) groupbyattrsprocessor: move metrics from OpenCensus to OpenTelemetry (#30763)
• (Contrib) loadbalancingexporter: Optimize metrics and traces export (#30141)
• (Contrib) all: Add component.UseLocalHostAsDefaultHost feature gate that changes default endpoints from 0.0.0.0 to localhost (#30702)
  This change affects the following components:
  • extension/health_check
  • receiver/jaeger
  • receiver/sapm
  • receiver/signalfx
  • receiver/splunk_hec
  • receiver/zipkin
• (Contrib) processor/resourcedetectionprocessor: Detect Azure cluster name from IMDS metadata (#26794)
• (Contrib) processor/transform: Add copy_metric function to allow duplicating a metric (#30846)

🧰 Bug fixes 🧰

• (Splunk) Fixes the value of a default environment variable used by Windows msi. (#4361)
• (Core) service: fix opencensus bridge configuration in periodic readers (#9361)
• (Core) otlpreceiver: Fix goroutine leak when GRPC server is started but HTTP server is unsuccessful (#9165)
• (Core) otlpexporter: PartialSuccess is treated as success, logged as warning. (#9243)
• (Contrib) basicauthextension: Accept empty usernames. (#30470)
  Per https://datatracker.ietf.org/doc/html/rfc2617#section-2, username and password may be empty strings ("").
  The validation used to enforce that usernames cannot be empty.
• (Contrib) pkg/ottl: Fix parsing of string escapes in OTTL (#23238)
• (Contrib) pkg/stanza: Recombine operator should always recombine partial logs (#30797)
  Previously, certain circumstances could result in partial logs being emitted without any
  recombiniation. This could occur when using is_first_entry, if the first partial log from
  a source was emitted before a matching "start of log" indicator was found. This could also
  occur when the collector was shutting down.
• (Contrib) pkg/stanza: Fix bug where recombine operator's 'overwrite_with' condition was inverted. (#30783)
• (Contrib) exporter/signalfx: Use "unknown" value for the environment correlation calls as fallback. (#31052)
  This fixed the APM/IM correlation in the Splunk Observability UI for the users that send traces with no "deployment.environment" resource attribute value set.

Docker Images:

• quay.io/signalfx/splunk-otel-collector:0.94.0-amd64 (digest: sha256:6d89f4d866af70e3d83880b355c8c3a3cac0c9bdabaf3ea9312250cf7c917649)
• quay.io/signalfx/splunk-otel-collector:0.94.0-arm64 (digest: sha256:15a090e07e61ec752471088e6db65f74540870c2942d162ecb866619fd7f9170)
• quay.io/signalfx/splunk-otel-collector:0.94.0-ppc64le (digest: sha256:2e640ccd1e46b9e707de7e991073baf1376ea8b2b869ec75bec0045a08018dab)
• quay.io/signalfx/splunk-otel-collector-windows:0.94.0 (digest: sha256:fd2683b996d6a0a79378576a705b0edc0a326df32c12cad87615307adf128a31)
• quay.io/signalfx/splunk-otel-collector-windows:0.94.0-2019 (digest: sha256:dec27aad01f6991e0535c37555682ae697827af2461fd14c09bd5e3cc7184efd)
• quay.io/signalfx/splunk-otel-collector-windows:0.94.0-2022 (digest: sha256:9b8cee1d6ba55bcbec2b280d1f23d6a69b556298d945618e62bca7e2fea6d22f)

v0.93.0

This Splunk OpenTelemetry Collector release includes changes from the opentelemetry-collector v0.93.0 and the opentelemetry-collector-contrib v0.93.0 releases where appropriate.

🛑 Breaking changes 🛑

• (Splunk) On Windows the SPLUNK_* environment variables were moved from the machine scope to the collector service scope This avoids collisions with other agents and instrumentation. If any of these environment variables are required by your apps, please adopt them directly. (#3930)
• (Splunk) mysql discovery now uses the OpenTelemetry Collector Contrib receiver by default instead of the smartagent receiver. (#4231)
• (Splunk) Stop sending internal Collector metrics from the batch processor. Drop them at the prometheus receiver level. (#4273)
• (Core) exporterhelper: remove deprecated exporterhelper.RetrySettings and exporterhelper.NewDefaultRetrySettings (#9256)
• (Contrib) vcenterreceiver: "receiver.vcenter.emitPerfMetricsWithObjects" feature gate is beta and enabled by default (#30615)
• (Contrib) docker: Adopt api_version as strings to correct invalid float truncation (#24025)
• (Contrib) extension/filestorage: Replace path-unsafe characters in component names (#3148)
  The feature gate extension.filestorage.replaceUnsafeCharacters is now enabled by default.
  See the File Storage extension's README for details.
• (Contrib) postgresqlreceiver: add feature gate receiver.postgresql.separateSchemaAttr to include schema as separate attribute (#29559)
  Enabling the featuregate adds a new resource attribute to store the schema of the table or index
  Existing table attributes are adjusted to not include the schema, which was inconsistently used

💡 Enhancements 💡

• (Splunk) Update opentelemetry-jmx-metrics version to 1.32.0 (#4201)
• (Core) configtls: add cipher_suites to configtls. (#8105)
  Users can specify a list of cipher suites to pick from. If left blank, a safe default list is used.
• (Core) service: mark telemetry.useOtelForInternalMetrics as stable (#816)
  (Splunk) Remove disabled telemetry.useOtelForInternalMetrics feature gate from our distribution. Some new internal metrics are now dropped at scrape time.
• (Core) exporters: Cleanup log messages for export failures (#9219)
  1. Ensure an error message is logged every time and only once when data is dropped/rejected due to export failure.
  2. Update the wording. Specifically, don't use "dropped" term when an error is reported back to the pipeline.
     Keep the "dropped" wording for failures happened after the enabled queue.
  3. Properly report any error reported by a queue. For example, a persistent storage error must be reported as a storage error, not as "queue overflow".
• (Contrib) pkg/stanza: Add a json array parser operator and an assign keys transformer. (#30321)
  Json array parser opreator can be used to parse a json array string input into a list of objects. |
  Assign keys transformer can be used to assigns keys from the configuration to an input list
• (Contrib) splunkhecexporter: Batch data according to access token and index, if present. (#30404)
• (Contrib) k8sattributesprocessor: Apply lifecycle tests to k8sprocessor, change its behavior to report fatal error (#30387)
• (Contrib) k8sclusterreceiver: add new disabled os.description, k8s.container_runtime.version resource attributes (#30342)
• (Contrib) k8sclusterreceiver: add os.type resource attribute (#30342)
• (Contrib) kubeletstatsreceiver: Add new *.cpu.usage metrics. (#25901)
• (Contrib) pkg/ottl: Add flatten function for flattening maps (#30455)
• (Contrib) redisreciever: adds metric for slave_repl_offset (#6942)
  also adds a shell script to set up docker-compose integration test
• (Contrib) receiver/sqlquery: Add debug log when running SQL query (#29672)

🧰 Bug fixes 🧰

• (Core) otlpreceiver: Ensure OTLP receiver handles consume errors correctly (#4335)
  Make sure OTLP receiver returns correct status code and follows the receiver contract (gRPC)
• (Core) zpagesextension: Remove mention of rpcz page from zpages extension (#9328)
• (Contrib) kafkareceiver: The Kafka receiver now exports some partition-specific metrics per-partition, with a partition tag (#30177)
  The following metrics now render per partition:
  • kafka_receiver_messages
  • kafka_receiver_current_offset
  • kafka_receiver_offset_lag

Docker Images:

• quay.io/signalfx/splunk-otel-collector:0.93.0-amd64 (digest: sha256:c2237acb476ce38263b4709326d9f9f3859d0f32de9ef8c5fde5bf6017ef8b39)
• quay.io/signalfx/splunk-otel-collector:0.93.0-arm64 (digest: sha256:52edfef95657a1058487456d8e5fe3fe9298fd65806b2c89b80ee1375c406f75)
• quay.io/signalfx/splunk-otel-collector:0.93.0-ppc64le (digest: sha256:8ee5a2979c84792aab5c9ea484e188f1614d9dd89fd0b3baa6c410200be72917)
• quay.io/signalfx/splunk-otel-collector-windows:0.93.0 (digest: sha256:77f91c139ca080d61253816e5929c4b88fe0ade88d51567481473cba56414900)
• quay.io/signalfx/splunk-otel-collector-windows:0.93.0-2022 (digest: sha256:c0e5c0870c199ec5f1c4b71aee61507abd67f9acd5defe3dad2511e78404ffa8)

v0.92.0

This Splunk OpenTelemetry Collector release includes changes from the opentelemetry-collector v0.92.0 and the opentelemetry-collector-contrib v0.92.0 releases where appropriate.

🛑 Breaking changes 🛑

• (Contrib) httpforwarder: Use confighttp.HTTPDefaultClientSettings when configuring the HTTPClientSettings for the httpforwarder extension. (#6641)
  By default, the HTTP forwarder extension will now use the defaults set in the extension:
  • The idle connection timeout is set to 90s.
  • The max idle connection count is set to 100.
• (Contrib) pkg/ottl: Now validates against extraneous path segments that a context does not know how to use. (#30042)
• (Contrib) pkg/ottl: Throw an error if keys are used on a path that does not allow them. (#30162)
• (Core) exporters/sending_queue: Do not re-enqueue failed batches, rely on the retry_on_failure strategy instead. (#8382)
  The current re-enqueuing behavior is not obvious and cannot be configured. It takes place only for persistent queue
  and only if retry_on_failure::enabled=true even if retry_on_failure is a setting for a different backoff retry
  strategy. This change removes the re-enqueuing behavior. Consider increasing retry_on_failure::max_elapsed_time
  to reduce chances of data loss or set it to 0 to keep retrying until requests succeed.
• (Core) confmap: Make the option WithErrorUnused enabled by default when unmarshaling configuration (#7102)
  The option WithErrorUnused is now enabled by default, and a new option WithIgnoreUnused is introduced to ignore
  errors about unused fields.

🚩 Deprecations 🚩

• (Contrib) k8sclusterreceiver: deprecate optional k8s.kubeproxy.version resource attribute (#29748)
• (Core) exporterhelper: Deprecate exporterhelper.RetrySettings in favor of configretry.BackOffConfig (#9091)
• (Core) extension/ballast: Deprecate memory_ballast extension. (#8343)
  Use GOMEMLIMIT environment variable instead.

💡 Enhancements 💡

• (Splunk) support core service validate command (#4175)
• (Splunk) Add routing connector to Splunk distribution (#4167)
• (Contrib) adopt splunkhec batch by token and index updates (#4151)
• (Contrib) vcenterreceiver: Add explicit statement of support for version 8 of ESXi and vCenter (#30274)
• (Contrib) routingconnector: routingconnector supports matching the statement only once (#26353)
• (Contrib) filterprocessor: Add telemetry for metrics, logs, and spans that were intentionally dropped via filterprocessor. (#13169)
• (Contrib) pkg/ottl: Add Hour OTTL Converter (#29468)
• (Contrib) kafkaexporter: add ability to publish kafka messages with message key of TraceID - it will allow partitioning of the kafka Topic. (#12318)
• (Contrib) kafkareceiver: Add three new metrics to record unmarshal errors. (#29302)
• (Contrib) hostmetricsreceiver: Add system.memory.limit metric reporting the total memory available. (#30306)
  This metric is opt-in. To enable it, set scrapers::memory::metrics::system.memory.limit::enabled to true in the hostmetrics config.
• (Contrib) kafkaexporter: Adds the ability to configure the Kafka client's Client ID. (#30144)
• (Contrib) pkg/stanza: Remove sampling policy from logger (#23801)
• (Contrib) resourcedetectionprocessor: Add "aws.ecs.task.id" attribute (#8274)
  Resourcedetectionprocessor now exports "aws.ecs.task.id" attribute, in addition to "aws.ecs.task.arn".
  This allows exporters like "awsemfexporter" to automatically pick up that attribute and make it available
  in templating (e.g. to use in CloudWatch log stream name).
• (Contrib) spanmetricsconnector: Fix OOM issue for spanmetrics by limiting the number of exemplars that can be added to a unique dimension set (#27451)
• (Contrib) connector/spanmetrics: Configurable resource metrics key attributes, filter the resource attributes used to create the resource metrics key. (#29711)
  This enhancement can be used to fix broken spanmetrics counters after a span producing service restart, when resource attributes contain dynamic/ephemeral values (e.g. process id).
• (Contrib) splunkhecreceiver: Returns json response in raw endpoint when it is successful (#29875)
• (Contrib) sqlqueryreceiver: Swap MS SQL Server driver from legacy 'denisenkom' to official Microsoft fork (#27200)
• (Core) exporterhelper: Add RetrySettings validation function (#9089)
  Validate that time.Duration, multiplier values in configretry are non-negative, and randomization_factor is between 0 and 1
• (Core) service: Enable telemetry.useOtelForInternalMetrics by updating the flag to beta (#7454)
  The metrics generated should be consistent with the metrics generated
  previously with OpenCensus. Splunk note: this option is disabled in our distribution. Users can enable the behaviour
  by setting --feature-gates +telemetry.useOtelForInternalMetrics at collector start if the new histograms are desired.
• (Core) confignet: Add dialer_timeout config option. (#9066)
• (Core) processor/memory_limiter: Update config validation errors (#9059)
  • Fix names of the config fields that are validated in the error messages
  • Move the validation from start to the initialization phrase
• (Core) exporterhelper: Add config Validate for TimeoutSettings (#9104)

🧰 Bug fixes 🧰

• (Contrib) filterset: Fix concurrency issue when enabling caching. (#11829)
• (Contrib) pkg/ottl: Fix issue with the hash value of a match subgroup in replace_pattern functions. (#29409)
• (Contrib) prometheusreceiver: Fix configuration validation to allow specification of Target Allocator configuration without providing scrape configurations (#30135)
• (Contrib) wavefrontreceiver: Return error if partially quoted (#30315)
• (Contrib) hosmetricsreceiver: change cpu.load.average metrics from 1 to {thread} (#29914)
• (Contrib) pkg/ottl: Fix bug where the Converter IsBool was not usable (#30151)
• (Contrib) time: The %z strptime format now correctly parses Z as a valid timezone (#29929)
  strptime(3) says that %z is "an RFC-822/ISO 8601 standard
  timezone specification", but the previous code did not allow the
  string "Z" to signify UTC time, as required by ISO 8601. Now, both
  +0000 and Z are recognized as UTC times in all components that
  handle strptime format strings.
• (Core) memorylimiterprocessor: Fixed leaking goroutines from memorylimiterprocessor (#9099)
• (Core) cmd/otelcorecol: Fix the code detecting if the collector is running as a service on Windows. (#7350)
  Removed the NO_WINDOWS_SERVICE environment variable given it is not needed anymore.
• (Core) otlpexporter: remove dependency of otlphttpreceiver on otlpexporter (#6454)

Docker Images:

• `quay.io/signalfx/splunk...

v0.91.3

🛑 Breaking changes 🛑

• (Splunk) - ecs-metadata sync the known_status property on the container_id dimension instead of lower cardinality container_name. This can be prevented by configuring dimensionToUpdate to container_name (#4091)
• (Splunk) Removes collectd/disk monitor (#3998)
  This monitor has been deprecated in favor of the disk-io monitor.
  Note that the disk-io monitor has a different dimension (disk
  instead of plugin_instance) to specify the disk.
• (Splunk) Removes collectd/df monitor (#3996)
  The monitor is deprecated and the filesystems monitor should be used instead.
• (Splunk) Removes netinterface monitor (#3991)
  This monitor is deprecated in favor of the net-io monitor.
• (Splunk) Removes collectd/vmem monitor (#3993)
  This monitor is deprecated in favor of the vmem monitor. The metrics should be fully compatible with this monitor.
• (Splunk) Removes collectd/load monitor (#3995)
  This monitor has been deprecated in favor of the load monitor. That monitor emits the same metrics and is fully compatible.
• (Splunk) Removes collectd/postgresql monitor (#3994)
  This monitor is deprecated in favor of the postgresql monitor.

💡 Enhancements 💡

• (Splunk) Adopt vcenter receiver (#4291)
• (Splunk) Adopt sshcheck receiver (#4099)
• (Splunk) Adopt awss3 exporter (#4117)
• (Splunk) Convert loglevel to verbosity on logging exporter (#4097)

Docker Images:

• quay.io/signalfx/splunk-otel-collector:0.91.3-amd64 (digest: sha256:d469e3372384604b36a6180525c9aacdd49fcbd082c618cfeede49fec7f35ada)
• quay.io/signalfx/splunk-otel-collector:0.91.3-arm64 (digest: sha256:2201612e13501157616eb2399d4474cbc91fcea5b36e2b0d4864b635f5c07c3a)
• quay.io/signalfx/splunk-otel-collector:0.91.3-ppc64le (digest: sha256:c967f7023d8bb8f260b63c8ec6de90c7ec4d3b87c010e743376fdfe5d6f5b92b)
• quay.io/signalfx/splunk-otel-collector-windows:0.91.3 (digest: sha256:05a736f642207dbfcc1f2c3ead2ff771b6ee8b746eec7925c4af62ee2c452a83)
• quay.io/signalfx/splunk-otel-collector-windows:0.91.3-2022 (digest: sha256:703f38c4431edefa43ed9b52ffcbbefe9463b9afb44443c5e845c6ec109d5d70)

v0.91.1

💡 Enhancements 💡

• (Splunk) Remove the project beta label (#4070)
• (Splunk) Source SPLUNK_LISTEN_INTERFACE on all host endpoints(#4065)
• (Splunk) Add support for start timestamps when using the light prometheus receiver (#4037)
• (Splunk) Node.js Auto Instrumentation:
  • Update splunk-otel-js to v2.6.0 (#4064)
  • Update linux installer script to use --global=false for local npm versions and configurations (#4068)

🛑 Breaking changes 🛑

• postgresql Discovery now uses the OpenTelemetry Collector Contrib receiver by default instead of the smartagent receiver (#3957)

Docker Images:

• quay.io/signalfx/splunk-otel-collector:0.91.1-amd64 (digest: sha256:86f99bf2224cbaa9ed5ebf7346e31619dc6d1ac16393dd089b635dc0606a55b0)
• quay.io/signalfx/splunk-otel-collector:0.91.1-arm64 (digest: sha256:0b15a2f8f0431f1b3e37060fed5d49a0aa661215c197f881aa22e992a1a255f7)
• quay.io/signalfx/splunk-otel-collector:0.91.1-ppc64le (digest: sha256:ee2898ba48c5ae41db268340388a12bfd12ad92a61f116fd62af6cad1d2a9097)
• quay.io/signalfx/splunk-otel-collector-windows:0.91.1 (digest: sha256:ca400b1cf29738d4cf4ce8012189b3e0b81072a875c239902c0ec0f503fa9237)
• quay.io/signalfx/splunk-otel-collector-windows:0.91.1-2022 (digest: sha256:ef6f41051f553b2766658d1f6a0cfbd69955fe342c21d2d4426089d9a27eeced)

v0.91.0

This Splunk OpenTelemetry Collector release includes changes from the opentelemetry-collector v0.91.0 and the opentelemetry-collector-contrib v0.91.0 releases where appropriate.

🛑 Breaking changes 🛑

• (Splunk) Node.js Auto Instrumentation:
  • The NODE_OPTIONS environment variable in the default config file has been updated to load the Node.js SDK from an absolute path (/usr/lib/splunk-instrumentation/splunk-otel-js/node_modules/@splunk/otel/instrument).
  • The Linux installer script now installs the Node.js SDK to /usr/lib/splunk-instrumentation/splunk-otel-js instead of globally.
  • The --npm-command Linux installer script option is no longer supported. To specify a custom path to npm, use the --npm-path <path> option.
• (Splunk) translatesfx: Remove translatesfx (#4028)
• (Splunk) collectd/elasticsearch: Remove collectd/elasticsearch monitor (#3997)

🚩 Deprecations 🚩

• (Splunk) collectd/cpu: Deprecate collectd/cpu explicitly. Please migrate to the cpu monitor (#4036)

💡 Enhancements 💡

• (Contrib) spanmetricsconnector: Add exemplars to sum metric (#27451)
• (Contrib) jaegerreceiver: mark featuregates to replace Thrift-gen with Proto-gen types for sampling strategies as stable (#27636)
  The following featuregate is stable:
  receiver.jaegerreceiver.replaceThriftWithProto
• (Contrib) kafkareceiver: Add the ability to consume logs from Azure Diagnostic Settings streamed through Event Hubs using the Kafka API. (#18210)
• (Contrib) resourcedetectionprocessor: Add detection of host.ip to system detector. (#24450)
• (Contrib) resourcedetectionprocessor: Add detection of host.mac to system detector. (#29587)
• (Contrib) pkg/ottl: Add silent ErrorMode to allow disabling logging of errors that are ignored. (#29710)
• (Contrib) postgresqlreceiver: Add config property for excluding specific databases from scraping (#29605)
• (Contrib) redisreceiver: Upgrade the redis library dependency to resolve security vulns in v7 (#29600)
• (Contrib) signalfxexporter: Enable HTTP/2 health check by default (#29716)
• (Contrib) splunkhecexporter: Enable HTTP/2 health check by default (#29717)
• (Contrib) statsdreceiver: Add support for 'simple' tags that do not have a defined value, to accommodate DogStatsD metrics that may utilize these. (#29012)
  This functionality is gated behind a new enable_simple_tags config boolean, as it is not part of the StatsD spec.
• (Core) service: add resource attributes as labels to otel metrics to ensures backwards compatibility with OpenCensus metrics. (#9029)
• (Core) config/confighttp: Exposes http/2 transport settings to enable health check and workaround golang http/2 issue golang/go#59690 (#9022)

🧰 Bug fixes 🧰

• (Splunk) migratecheckpoint: Migrating offsets from SCK to SCK-Otel doesn't work. This is because of incorrect keys we use to populate the boltdb cache. (#3879)
• (Contrib) connector/spanmetrics: Fix memory leak when the cumulative temporality is used. (#27654)
• (Contrib) splunkhecexporter: Do not send null event field values in HEC events. Replace null values with an empty string. (#29551)
• (Contrib) k8sobjectsreceiver: fix k8sobjects receiver fails when some unrelated Kubernetes API is down (#29706)
• (Contrib) resourcedetectionprocessor: Change type of host.cpu.model.id and host.cpu.model.family from int to string. (#29025)
  Disable the processor.resourcedetection.hostCPUModelAndFamilyAsString feature gate to get the old behavior.
• (Contrib) filelogreceiver: Fix problem where checkpoints could be lost when collector is shutdown abruptly (#29609, #29491)
• (Contrib) pkg/stanza: Allow key_value_parser to parse values that contain the delimiter string. (#29629
• (Core) exporterhelper: fix missed metric aggregations (#9048)
  This ensures that context cancellation in the exporter doesn't interfere with metric aggregation. The OTel
  SDK currently returns if there's an error in the context used in Add. This means that if there's a
  cancelled context in an export, the metrics are now recorded.

Docker Images:

• quay.io/signalfx/splunk-otel-collector:0.91.0-amd64 (digest: sha256:6ea26505ebc2aa46a9a30d183b6a6aa98333ba29a28c6a1ca4c64137665f5bfc)
• quay.io/signalfx/splunk-otel-collector:0.91.0-arm64 (digest: sha256:810949a8c9ddfebce8e519084f594f93d2688dd70e1b22714c93bb5e765e20d4)
• quay.io/signalfx/splunk-otel-collector:0.91.0-ppc64le (digest: sha256:147dc913bcc0de535fed048e4914d9fd9521d0db10ce838199530c3b4041fd27)
• quay.io/signalfx/splunk-otel-collector-windows:0.91.0 (digest: sha256:9179df80646b699f294bb07a199cf49d4f10206676b6dec9dec340e26c9cdbdd)
• quay.io/signalfx/splunk-otel-collector-windows:0.91.0-2022 (digest: sha256:893155430652af699c80364e69212b3c8957ca8859d34d1ab629e24630354cff)

v0.90.0

This Splunk OpenTelemetry Collector release includes changes from the opentelemetry-collector v0.90.1 and the opentelemetry-collector-contrib v0.90.0 releases where appropriate.

🛑 Breaking changes 🛑

• (Core) service: To remain backwards compatible w/ the metrics generated today, otel generated metrics will be generated without the _total suffix (#7454)
• (Core) service: use WithNamespace instead of WrapRegistererWithPrefix (#8988)
  Using this functionality in the otel prom exporter fixes a bug where the
  target_info was prefixed as otelcol_target_info previously.

🚩 Deprecations 🚩

• (Splunk) Deprecate collectd/marathon (#3992)
• (Splunk) Add deprecation notice to collectd/etcd (use etcd instead) (#3990)
• (Splunk) Mark translatesfx as deprecated (#3984)

💡 Enhancements 💡

• (Splunk) mysqlreceiver: Add mysqlreceiver to the Splunk distribution (#3989)
• (Core) exporter/debug: Change default verbosity from normal to basic (#8844)
  This change has currently no effect, as basic and normal verbosity share the same behavior. This might change in the future though, with the normal verbosity being more verbose than it currently is (see open-telemetry/opentelemetry-collector#7806). This is why we are changing the default to basic, which is expected to stay at the current level of verbosity (one line per batch).
• (Core) exporterhelper: Fix shutdown logic in persistent queue to not require consumers to be closed first (#8899)
• (Core) confighttp: Support proxy configuration field in all exporters that support confighttp (#5761)
• (Contrib) resourcedetectionprocessor: Add k8s cluster name detection when running in EKS (#26794)
• (Contrib) pkg/ottl: Add new IsDouble function to facilitate type checking. (#27895)
• (Contrib) mysqlreceiver: expose tls in mysqlreceiver (#29269)
  If tls is not set, the default is to disable TLS connections.
• (Contrib) processor/transform: Convert between sum and gauge in metric context when alpha feature gate processor.transform.ConvertBetweenSumAndGaugeMetricContext enabled (#20773)
• (Contrib) receiver/mongodbatlasreceiver: adds project config to mongodbatlas metrics to filter by project name and clusters. (#28865)
• (Contrib) pkg/stanza: Add "namedpipe" operator. (#27234)
• (Contrib) pkg/resourcetotelemetry: Do not clone data in pkg/resourcetotelemetry by default (#29327)
  The resulting consumer will be marked as MutatesData instead
• (Contrib) pkg/stanza: Improve performance by not calling decode when nop encoding is defined (#28899)
• (Contrib) receivercreator: Added support for discovery of endpoints based on K8s services (#29022)
  By discovering endpoints based on K8s services, a dynamic probing of K8s service leveraging for example the httpcheckreceiver get enabled
• (Contrib) signalfxexporter: change default timeout to 10 seconds (#29436)
• (Contrib) hostmetricsreceiver: Add optional Linux-only metric system.linux.memory.available (#7417)
  This is an alternative to system.memory.usage metric with state=free.
  Linux starting from 3.14 exports "available" memory. It takes "free" memory as a baseline, and then factors in kernel-specific values.
  This is supposed to be more accurate than just "free" memory.
  For reference, see the calculations here.
  See also MemAvailable in /proc/meminfo.

🧰 Bug fixes 🧰

• (Splunk) cmd/otelcol: Fix the code detecting if the collector is running as a service on Windows. The fix should make
  setting the NO_WINDOWS_SERVICE environment variable unnecessary. (#4002)
• (Core) exporterhelper: Fix invalid write index updates in the persistent queue (#8115)
• (Contrib) filelogreceiver: Fix issue where files were unnecessarily kept open on Windows (#29149)
• (Contrib) mongodbreceiver: add receiver.mongodb.removeDatabaseAttr Alpha feature gate to remove duplicate database name attribute (#24972)
• (Contrib) pkg/stanza: Fix panic during stop for udp async mode only. (#29120)

Docker Images:

• quay.io/signalfx/splunk-otel-collector:0.90.0-amd64 (digest: sha256:7368c1f88063851f48f92a46150955fef943420dc7e5fa0bbd2f443e2352bac7)
• quay.io/signalfx/splunk-otel-collector:0.90.0-arm64 (digest: sha256:6967fe7a2910eea0e0c53dba3a51517df5392142a1a4f0c762f4c5368b0f09f9)
• quay.io/signalfx/splunk-otel-collector:0.90.0-ppc64le (digest: sha256:6a1dd1b9e480c8baa83d93507446954e13bd11ee754b8534b0b4dddd307c23ea)
• quay.io/signalfx/splunk-otel-collector-windows:0.90.0 (digest: sha256:4cd076173486ac09d61d3c306ef297e07f77eacd08b485b3737bc7bed9eb78e8)
• quay.io/signalfx/splunk-otel-collector-windows:0.90.0-2022 (digest: sha256:89456a69cec86cd9e391b7809373ef10738a1dbe184167ab64cf0052ec3ca159)

v0.89.0

🛑 Breaking changes 🛑

• (Contrib) pkg/stanza/receiver/windowseventlog: Improve parsing of Windows Event XML by handling anonymous Data elements. (#21491)
  This improves the contents of Windows log events for which the publisher manifest is unavailable. Previously, anonymous Data elements were ignored. This is a breaking change for users who were relying on the previous data format.
• (Contrib) processor/k8sattributes: Graduate "k8sattr.rfc3339" feature gate to Beta. (#28817)
  Time format of k8s.pod.start_time attribute value migrated from RFC3339:
  Before: 2023-07-10 12:34:39.740638 -0700 PDT m=+0.020184946
  After: 2023-07-10T12:39:53.112485-07:00
  The feature gate can be temporary reverted back by adding --feature-gate=-k8sattr.rfc3339 to the command line.
• (Contrib) receiver/filelogreceiver: Change "Started watching file" log behavior (#28491)
  Previously, every unique file path which was found by the receiver would be remembered indefinitely.
  This list was kept independently of the uniqueness / checkpointing mechanism (which does not rely on the file path).
  The purpose of this list was to allow us to emit a log whenever a path was seen for the first time.
  This removes the separate list and relies instead on the same mechanism as checkpointing. Now, a similar log is emitted
  any time a file is found which is not currently checkpointed. Because the checkpointing mechanism does not maintain history
  indefinitely, it is now possible that a log will be emitted for the same file path. This will happen when no file exists at
  the path for a period of time.

🚩 Deprecations 🚩

• (Contrib) postgresqlreceiver: Deprecation of postgresql replication lag metrics postgresql.wal.lag in favor of more precise 'postgresql.wal.delay' (#26714)

💡 Enhancements 💡

• (Splunk) receiver/mongodbreceiver: Adds mongobdreceiver in Splunk collector distro (#3979)
• (Contrib) processor/tailsampling: adds optional upper bound duration for sampling (#26115)
• (Contrib) collectdreceiver: Add support of confighttp.HTTPServerSettings (#28811)
• (Contrib) collectdreceiver: Promote collectdreceiver as beta component (#28658)
• (Contrib) receiver/hostmetricsreceiver: Added support for host's cpuinfo frequnecies. (#27445)
  In Linux the current frequency is populated using the values from /proc/cpuinfo. An os specific implementation will be needed for Windows and others.
• (Contrib) receiver/hostmetrics/scrapers/process: add configuration option to mute error reading username for process (#14311, #17187)
• (Contrib) azureevenhubreceiver: Allow the Consumer Group to be set in the Configuration. (#28633)
• (Contrib) spanmetricsconnector: Add Events metric to span metrics connector that adds list of event attributes as dimensions (#27451)
• (Contrib) processor/k8sattribute: support adding labels and annotations from node (#22620)
• (Contrib) windowseventlogreceiver: Add parsing for Security and Execution event fields. (#27810)
• (Contrib) filelogreceiver: Add the ability to order files by mtime, to only read the most recently modified files (#27812)
• (Contrib) wavefrontreceiver: Wrap metrics receiver under carbon receiver instead of using export function (#27248)
• (Contrib) pkg/ottl: Add IsBool function into OTTL (#27897)
• (Contrib) k8sclusterreceiver: add k8s.node.condition metric (#27617)
• (Contrib) kafkaexporter/kafkametricsreceiver/kafkareceiver: Expose resolve_canonical_bootstrap_servers_only config (#26022)
• (Contrib) receiver/mongodbatlasreceiver: Enhanced collector logs to include more information about the MongoDB Atlas API calls being made during logs retrieval. (#28851)
• (Contrib) receiver/mongodbatlasreceiver: emit resource attributes "mongodb_atlas.region.name" and "mongodb_atlas.provider.name" on metric scrape. (#28833)
• (Contrib) processor/resourcedetection: Add processor.resourcedetection.hostCPUModelAndFamilyAsString feature gate to change the type of host.cpu.family and host.cpu.model.id attributes from int to string. (#29025)
  This feature gate will graduate to beta in the next release.
• (Contrib) processor/tailsampling: Optimize performance of tailsamplingprocessor (#27889)
• (Contrib) redisreceiver: include server.address and server.port resource attributes (#22044)
• (Contrib) spanmetricsconnector: Add exemplars to sum metric (#27451)
• (Core) service/extensions: Allow extensions to declare dependencies on other extensions and guarantee start/stop/notification order accordingly. (#8732)
• (Core) exporterhelper: Log export errors when retry is not used by the component. (#8791)

🧰 Bug fixes 🧰

• (Splunk) smartagent/processlist: Reduce CPU usage when collecting process information on Windows (#3980)
• (Contrib) filelogreceiver: Fix issue where counting number of logs emitted could cause panic (#27469, #29107)
• (Contrib) kafkareceiver: Fix issue where counting number of logs emitted could cause panic (#27469, #29107)
• (Contrib) k8sobjectsreceiver: Fix issue where counting number of logs emitted could cause panic (#27469, #29107)
• (Contrib) fluentforwardreceiver: Fix issue where counting number of logs emitted could cause panic (#27469, #29107)
• (Contrib) azureeventhubreceiver: Updated documentation around Azure Metric to OTel mapping. (#28622)
• (Contrib) receiver/hostmetrics: Fix panic on load_scraper_windows shutdown (#28678)
• (Contrib) splunkhecreceiver: Do not encode JSON response objects as string. (#27604)
• (Contrib) processor/k8sattributes: Set attributes from namespace/node labels or annotations even if node/namespaces name attribute are not set. (#28837)
• (Contrib) pkg/stanza: Fix data-corruption/race-condition issue in udp async (reuse of buffer); use buffer pool instead. (#27613)
• (Contrib) zipkinreceiver: Return BadRequest in case of permanent errors (#4335)
• (Core) exporterhelper: fix bug with queue size and capacity metrics (#8682)

Docker Images:

• quay.io/signalfx/splunk-otel-collector:0.89.0-amd64 (digest: sha256:a9c3ff227f6926ed1904ba143206e08820be093fa38271f24f0b05383cc6273b)
• quay.io/signalfx/splunk-otel-collector:0.89.0-arm64 (digest: sha256:02e07363cb15311a70588074213817c311250361c0a470b25cb9c7813428cf96)
• quay.io/signalfx/splunk-otel-collector:0.89.0-ppc64le (digest: sha256:3f2064347852a3e28b10fbcb1a6bc99276a860b6953fd9c7b4f77b340e5b4e7c)
• quay.io/signalfx/splunk-otel-collector-windows:0.89.0 (digest: `sha256:c6f1965c2feb5c2db611...

v0.88.0

This Splunk OpenTelemetry Collector release includes changes from the opentelemetry-collector v0.88.0 and the opentelemetry-collector-contrib v0.88.0 releases where appropriate.

🛑 Breaking changes 🛑

• (Splunk) smartagent: Respect JAVA_HOME environment variable instead of enforcing bundle-relative value (#3877)
• (Contrib) k8sclusterreceiver: Remove opencensus.resourcetype resource attribute (#26487)
• (Contrib) splunkhecexporter: Remove max_connections configuration setting. (#27610)
  • use max_idle_conns or max_idle_conns_per_host instead.
• (Contrib) signalfxexporter: Remove max_connections configuration setting. (#27610)
  • use max_idle_conns or max_idle_conns_per_host instead.
• (Core) exporterhelper: make enqueue failures available for otel metrics (#8673). This will prevent internal Collector otelcol_exporter_enqueue_failed_<telemetry_type> metrics from being reported unless greater than 0.

💡 Enhancements 💡

• (Splunk) Add an option, -msi_public_properties, to allow passing MSI public properties when installing the Splunk OpenTelemetry Collector using the Windows installer script (#3921)
• (Splunk) Add support for config map providers in discovery configuration. (#3874)
• (Splunk) Add zero config support for chef deployments (#3903)
• (Splunk) Add zero config support for puppet deployments (#3922)
• (Contrib) receiver/prometheus: Warn instead of failing when users rename using metric_relabel_configs in the prometheus receiver (#5001)
• (Contrib) k8sobjectsreceiver: Move k8sobjectsreceiver from Alpha stability to Beta stability for logs. (#27635)
• (Contrib) doubleconverter: Adding a double converter into pkg/ottl (#22056)
• (Contrib) syslogreceiver: validate protocol name (#27581)
• (Contrib) entension/storage/filestorage: Add support for setting bbolt fsync option (#20266)
• (Contrib) filelogreceiver: Add a new "top_n" option to specify the number of files to track when using ordering criteria (#23788)
• (Contrib) k8sclusterreceiver: add optional k8s.pod.qos_class resource attribute (#27483)
• (Contrib) pkg/stanza: Log warning, instead of error, when Windows Event Log publisher metadata is not available and cache the successfully retrieved ones. (#27658)
• (Contrib) pkg/ottl: Add optional Converter parameters to replacement Editors (#27235)
• (Contrib) signalfxexporter: Add an option to control the dimension client timeout (#27815)
• (Contrib) signalfxexporter: Add the build version to the user agent of the SignalFx exporter (#16841)

🧰 Bug fixes 🧰

• (Splunk) Fix Tanzu Tile to properly set proxy exclusions. (#3902)
• (Contrib) syslog: add integration tests and fix related bugs (#21245)
• (Contrib) processor/resourcedetection: Don't parse the field cpuInfo.Model if it's blank. (#27678)
• (Contrib) k8sclusterreceiver: Change clusterquota and resourcequota metrics to use {resource} unit (#10553)
• (Contrib) pkg/ottl: Fix bug where named parameters needed a space after the equal sign (=). (#28511)
• (Contrib) filelogreceiver: Fix issue where batching of files could result in ignoring start_at setting. (#27773)
• (Core) exporterhelper: Fix nil pointer dereference when stopping persistent queue after a start encountered an error (#8718)

💡 Enhancements 💡

• (Splunk) Add an option, -msi_public_properties, to allow passing MSI public properties when installing the Splunk OpenTelemetry Collector using the Windows installer script (#3921)

Docker Images:

• quay.io/signalfx/splunk-otel-collector:0.88.0-amd64 (digest: sha256:f458b63c48e18a5deb4fba47fb49a2fe0dace60948d99f7d57d4443824d856aa)
• quay.io/signalfx/splunk-otel-collector:0.88.0-arm64 (digest: sha256:36432be420720026ec8b95fd5d2c24c24ba54a64c02a87f6c121ce1aaeec26f0)
• quay.io/signalfx/splunk-otel-collector:0.88.0-ppc64le (digest: sha256:28bc1be941f59d29dedf145afa758fcb0dce64dc0a3cfce5e73bef8fc2cb1ac2)
• quay.io/signalfx/splunk-otel-collector-windows:0.88.0 (digest: sha256:3997d7fe7bcbb4f76da1c4f856358b50724e1002ffb4df2e92377a7f47912a35)
• quay.io/signalfx/splunk-otel-collector-windows:0.88.0-2022 (digest: sha256:542a9487988d80a1e8b6e3874398590953be0b17c511fa5c4728128e079904c7)

v0.87.0

This Splunk OpenTelemetry Collector release includes changes from the opentelemetry-collector v0.87.0 and the opentelemetry-collector-contrib v0.87.0 releases where appropriate.

🛑 Breaking changes 🛑

• (Splunk) Auto Instrumentation for Linux (#3791):
  • The /usr/lib/splunk-instrumentation/instrumentation.conf config file is no longer
    supported, and is replaced by /etc/splunk/zeroconfig/java.conf. If the splunk-otel-auto-instrumentation deb/rpm
    package is manually upgraded, the options within /usr/lib/splunk-instrumentation/instrumentation.conf will need to
    be manually migrated to their corresponding environment variables within /etc/splunk/zeroconfig/java.conf.
  • Manual installation of the splunk-otel-auto-instrumentation deb/rpm package no longer automatically adds
    /usr/lib/splunk-instrumentation/libsplunk.so to /etc/ld.so.preload.
  • Manual upgrade of the splunk-otel-auto-instrumentation deb/rpm package will automatically remove
    /usr/lib/splunk-instrumentation/libsplunk.so from /etc/ld.so.preload.
  • The splunk.linux-autoinstr.executions metric is no longer generated by libsplunk.so.
  • See Splunk OpenTelemetry Zero Configuration Auto Instrumentation for Linux
    for manual installation/configuration details.
  • For users of the Ansible, Chef, Puppet, or Salt modules for Auto Instrumentation, it is recommended to update the following option in your configuration for version 0.86.0 or older until these modules are updated to manage these changes:
    • Ansible: splunk_otel_auto_instrumentation_version
    • Chef: auto_instrumentation_version
    • Puppet: auto_instrumentation_version
    • Salt: auto_instrumentation_version
• (Contrib) kubeletstatsreceiver: Fixes a bug where the "insecure_skip_verify" config was not being honored when "auth_type" is "serviceAccount" in kubelet client. (#26319)
  • Before the fix, the kubelet client was not verifying kubelet's certificate. The default value of the config is false,
    so with the fix the client will start verifying tls cert unless the config is explicitly set to true.
• (Contrib) tailsamplingprocessor: Improve counting for the count_traces_sampled metric (#25882)
• (Contrib) extension/filestorage: Replace path-unsafe characters in component names (#3148)
• (Core) service/telemetry exporter/exporterhelper: Enable sampling logging by default and apply it to all components. (#8134)
  • The sampled logger configuration can be disabled easily by setting the service::telemetry::logs::sampling::enabled to false.

🚩 Deprecations 🚩

• (Splunk) The following Auto Instrumentation options for the Linux installer script are deprecated and will only apply if the --instrumentation-version <version> option is specified for version 0.86.0 or older:
  • --[no-]generate-service-name: libsplunk.so no longer generates service names for instrumented applications. The default behavior is for the activated Java and/or Node.js Auto Instrumentation agents to automatically generate service names. Use the --service-name <name> option to override the auto-generated service names for all instrumented applications.
  • --[enable|disable]-telemetry: libsplunk.so no longer generates the splunk.linux-autoinstr.executions telemetry metric.

🚀 New components 🚀

• (Splunk) Add the loadbalancing exporter (#3825)
• (Splunk) Add the udplog receiver (#3826)

💡 Enhancements 💡

• (Splunk) Update golang to 1.20.10 (#3770)
• (Splunk) Add debian 12 support to installer (#3766)
• (Splunk) Add new Auto Instrumentation options for the Linux installer script (#3791):
  • --with[out]-systemd-instrumentation: Activate auto instrumentation for only systemd services without preloading
    the libsplunk.so shared object library (default: --without-systemd-instrumentation)
  • Initial support for Splunk OpenTelemetry Auto Instrumentation for Node.js:
    • Activated by default if the --with-instrumentation or --with-systemd-instrumentation option is specified.
    • Use the --without-instrumentation-sdk node option to explicitly skip Node.js.
    • npm is required to install the Node.js Auto Instrumentation package. If npm is not installed, Node.js will
      be skipped automatically.
    • By default, the Node.js Auto Instrumentation package is installed with the npm install --global command. Use the
      --npm-command "<command>" option to specify a custom command.
    • Environment variables to activate and configure Node.js auto instrumentation are added to /etc/splunk/zeroconfig/node.conf (for --with-instrumentation) or
      /usr/lib/systemd/system.conf.d/00-splunk-otel-auto-instrumentation.conf (for --with-systemd-instrumentation) based on defaults and specified installation options.
  • Auto Instrumentation for Java is also activated by default if the --with-instrumentation or
    --with-systemd-instrumentation option is specified. Use the --without-instrumentation-sdk java option to skip Java.
  • --otlp-endpoint host:port: Set the OTLP gRPC endpoint for captured traces (default: http://LISTEN_INTERFACE:4317
    where LISTEN_INTERFACE is the value from the --listen-interface option if specified, or 127.0.0.1 otherwise)
  • See Linux Installer Script for more details.
• (Splunk) Update splunk-otel-javaagent to v1.29.0 (#3788)
• (Splunk) Redis discovery (#3731)
• (Splunk) Update Bundled OpenJDK to 11.0.21+9 (#3819)
• (Splunk) Oracledb discovery tweaks (remove static endpoint) (#3836)
• (Contrib) probabilisticsamplerprocessor: Allow non-bytes values to be used as the source for the sampling decision (#18222)
• (Contrib) kafkareceiver: Allow users to attach kafka header metadata with the log/metric/trace record in the pipeline. Introduce a new config param, 'header_extraction' and some examples. (#24367)
• (Contrib) kafkaexporter: Adding Zipkin encoding option for traces to kafkaexporter (#21102)
• (Contrib) kubeletstatsreceiver: Support specifying context for kubeConfig auth_type (#26665)
• (Contrib) kubeletstatsreceiver: Adds new k8s.pod.cpu_limit_utilization, k8s.pod.cpu_request_utilization, k8s.container.cpu_limit_utilization, and k8s.container.cpu_request_utilization metrics that represent the ratio of cpu used vs set limits and requests. (#27276)
• (Contrib) kubeletstatsreceiver: Adds new k8s.pod.memory_limit_utilization, k8s.pod.memory_request_utilization, k8s.container.memory_limit_utilization, and k8s.container.memory_request_utilization metrics that represent the ratio of memory used vs set limits and requests. (#25894)

🧰 Bug fixes 🧰

• (Contrib) spanmetricsprocessor: Prune histograms when dimension cache is pruned. (#27080)
  • Dimension cache was always pruned but histograms were not being pruned. This caused metric series created
    by processor/spanmetrics to grow unbounded.
• (Contrib) splunkhecreceiver: Fix receiver behavior when used for metrics and logs at the same time; metrics are no longer dropped. (#27473)
• (Contrib) metricstransformprocessor: Fixes a nil pointer dereference when copying an exponential histogram (#27409)
• (contrib) k8sclusterreceiver: change k8s.container.ready, k8s.pod.phase, k8s.pod.status_reason, k8s.namespace.phase units to empty (#10553)
• (...