Releases: DataDog/lading
Releases · DataDog/lading
v0.24.0
Release lading 0.24.0 This release modifies the logrotate FS configuration to include a load profile, allowing the user to specify either constant load for a linearly increasing load. This is a breaking configuration change. Signed-off-by: Brian L. Troutwine <[email protected]>
v0.23.5
Fixed
- Target pid watcher will not report 0 for containers.
Removed
- Container is now the only supported release artifact, removing crates.io
publish and binaries.
v0.23.4
Added
- Introduced logrotate_fs, a sub-generator of
file_gen
that exposes a FUSE
filesystem to mimic log rotation. Accurately records bytes lost by readers via
rotation.
Fixed
- The target metrics prometheus parser now handles labels that have spaces in
them rather than incorrectly identifying the metric value. - Prometheus target metrics scraper will no longer panic if a metric has an
invalid value (instead it will be logged) - HTTP traffic sent by the HTTP generator and Splunk generator now always includes
a Host header to comply with HTTP 1.1 requirements
v0.23.3
Changed
- Linux observer is more resilient to scenarios where lading lacks ptrace permission.
Removed
- lading_capture no longer exports a protobuf version of the capture.
Added
- HTTP blackhole now has a
response_delay_millis
setting, allowing for
simulation of latent network connections. The default value is 0.
Fixed
- Logrotate file generator will no longer panic when rotating files beyond one depth.
v0.23.2
Changed
- Now built using rust 1.81.0.
Fixed
- Warmup period is now respected when container targeting is in use.
- Capture manager waits for the target to start running before recording data.
v0.23.2-rc2
Changed
- Now built using rust 1.81.0.
Fixed
- Warmup period is now respected when container targeting is in use.
- Capture manager waits for the target to start running before recording data.
v0.23.2-rc1
Changed
- Now built using rust 1.81.0.
Fixed
- Warmup period is now respected when container targeting is in use.
v0.23.1
Fixed
- Fixes a panic in the signal mechanism that appeared when using the file
generator most prominately.
v0.23.0
Added
- Added ability to create tags for both expvar and prometheus target metrics specific to a single target_metrics configuration (example below shows prometheus metrics collected from the core agent and two additional tags created)
target_metrics: - prometheus: #core agent telemetry uri: "http://127.0.0.1:5000/telemetry" tags: sub_agent: "core" any_label: "any-string-value"
v0.23.0-rc0
Added
- Added ability to create tags for both expvar and prometheus target metrics specific to a single target_metrics configuration (example below shows prometheus metrics collected from the core agent and two additional tags created)
target_metrics: - prometheus: #core agent telemetry uri: "http://127.0.0.1:5000/telemetry" tags: sub-agent: "core-agent" any-label: "any-string-value"