Skip to content

Releases: cloudfoundry/metric-store-release

Metric Store 1.5.4

09 Aug 15:11
Compare
Choose a tag to compare
  • Added support for TAS v2.12.

Metric Store 1.5.1

16 Nov 21:58
Compare
Choose a tag to compare

Changelog:

  • Include and use route registration health check.
  • Enable replication in tile version of metric-store. Metrics stored prior to this version without replication will be accessible, however they will not be replicated. Only metrics egressed after this upgrade will be replicated. This will also double previous storage requirements and may return inconsistent query results for metrics stored before the upgrade
  • Update tile internal certs to include missing SANs.

Metric Store 1.5.0

16 Oct 21:58
Compare
Choose a tag to compare

NOTE: this was never officially released, v1.5.1 is recommended instead

Changelog:

  • Enable replication in tile version of metric-store. Metrics stored prior to this version without replication will be accessible, however they will not be replicated. Only metrics egressed after this upgraded will be replicated.
  • Update tile internal certs to include missing SANs. These certs are regenerated automatically
  • Include and use route registration health check.

Metric Store 1.4.4

17 Sep 20:18
Compare
Choose a tag to compare

Changelog:
Certs:

  • Interpret common name as hostname for certs until certs get updated [https://www.pivotaltracker.com/story/show/174810994]
  • regenerate testing certificates using SANs
  • Use opsman ca certificate as the proxy ca certificate [https://www.pivotaltracker.com/story/show/173816208]
  • use properties to create tile mTLS certs [https://www.pivotaltracker.com/story/show/173816208]
  • use ops manager ca cert to sign metric store egress certificates [https://www.pivotaltracker.com/story/show/173816208]

HTTP Rollup:

  • Add instance_id to tagged rollup http metrics [https://www.pivotaltracker.com/story/show/174363898]
  • only emit http metrics that were seen in the rollup interval fixes [https://www.pivotaltracker.com/story/show/173421191], rolls back feature for [https://www.pivotaltracker.com/story/show/172614969]

Integrations:

  • Remove healthwatch scraping and config [https://www.pivotaltracker.com/story/show/174637412]
  • Add bosh-dns-aliases-release to TAS tile fixes github issue #102, [https://www.pivotaltracker.com/story/show/173785442]

Stability:

  • Disable concurrent queries by default [https://www.pivotaltracker.com/story/show/173775331]
  • Promote query concurrency option to spec [https://www.pivotaltracker.com/story/show/173775331]
  • Remove thread unsafe buffer from tcp client [https://www.pivotaltracker.com/story/show/173421191]
  • log all instances of metric_store_dropped_points_total [https://www.pivotaltracker.com/story/show/173421191]

Cleanup:

  • Remove smoke-tests bosh job, since it does nothing

Metric Store 1.4.3

14 Jul 17:34
Compare
Choose a tag to compare

Changelog:

  • expose metrics and debug endpoints on different ports
  • use context to control child and parent go routines in leanstreams
  • improve self-scraping
  • update major dependencies: Go 1.13.12, Prometheus 2.19.1

Metric Store 1.4.2

23 Apr 21:44
Compare
Choose a tag to compare

Changelog:

  • provide missing NODE_INDEX in metric-store-nozzle job

Metric Store 1.4.1

13 Apr 21:23
Compare
Choose a tag to compare

Changelog:

  • opt into the Ops Manager provided syslog form
  • allow inline certificates for alert managers
  • add a PKS tile
  • update major dependencies: Go 1.13.9

Metric Store 1.4.0

20 Mar 20:41
Compare
Choose a tag to compare

Changelog:

  • BREAKING CHANGE: API change to allow complete configuration of alertmanager via API
  • bump crypto/ssh dependency due to vulnerability
  • remove configuration of alertmanager via ENV vars

Metric Store 1.3.0

25 Feb 23:12
Compare
Choose a tag to compare

Changelog:

  • BREAKING CHANGE: (if upgrading from private multi-node) routing bug fix
  • add replication across a multi-node cluster (merge in previously private multi-node project)
  • add support for custom prometheus recording and alerting rules api
  • provide go client for rules API
  • add blackbox process which optionally monitors metric-store and provides key SLI/SLOs
  • add prometheus style scraping of metric endpoints
  • add indicator protocol support
  • increase read sample limit 20 million (from 1 million)
  • secure metrics endpoints with mTLS
  • remove use of PromScraper to inject metrics into loggregator
  • bump Prometheus: 2.15.2

Metric Store 1.2.2

19 Nov 21:36
Compare
Choose a tag to compare

Changelog:

  • improve timer rollup metrics to handle old and new gorouter versions