v1.3.4

Service Telemetry Operator v1.3.4

Minor changes updating Grafana resources to work with new eventing data model

Changes since v.1.3.3
160ed89 (HEAD -> stable-1.3, origin/stable-1.3) adjust elasticsearch timefield to exist with new event data model (#232) (#234)

v1.3.3

Service Telemetry Operator v1.3.3

Minor changes around testing and release delivery.

Changes since v1.3.2

9d30e97 (HEAD -> stable-1.3, origin/stable-1.3) Adjust bundle Dockerfile to build for a bundle (#228) (#230)
c3ff697 Adjust sensubility telemetry bus topic (#226) (#227)

v1.3.2

Service Telemetry Operator v1.3.2

Provides fixes for releasing against OCP v4.7 and new Smart Gateway deployment for sensubility metrics.

Changes since v1.3.1

7737518 (HEAD -> stable-1.3, origin/stable-1.3) Deploy a sensubility smartgateway (#224) (#225)
8a1190b Operators supporting 4.6+ need backport=false (#222) (#223)

v1.3.1

Service Telemetry Operator v1.3.1

Minor changes primarily supporting bundles.

Changes since v1.3.0

20bf87b (HEAD -> stable-1.3, origin/stable-1.3) Bump OCP installable version to include v4.7 (#219) (#220)
7cee0a8 Add port 5672 back to listeners for Interconnect (#218)
095f005 Explicitly specify namespace for each API reference (#212) (#215)
c96ae91 Break out common configuration to enable-stf.yaml (#211) (#213)
30380c0 Adjust grafana datasource name to be more general (#207) (#209)
248fae4 Allow override path to operator-sdk binary (#206)
e9be6a2 Fix invalid olm.skipRelease with olm.skipRange (#204) (#205)
a1aba29 Allow better controls when generating bundles (#202) (#203)

v1.3.0

Service Telemetry Operator v1.3.0

Major release of the Service Telemetry Operator providing support for OCP 4.6, updated stf-run-ci role for automated deployment for development, and support for the refactored Smart Gateway Operator. Also adds scripts for generating Operator bundles dynamically from templated files. Telemetry data from Ceilometer now shows up as a single metric from a single publisher by dropping the publisher label, allowing for better support when OpenStack is running with multiple controllers.

This release drops support for infra.watch/v1alpha1 CRD interface and only supports v1beta1 and adds support for smartgateway.infra.watch/v2.

Changes since v1.2.1

248fae4 Allow override path to operator-sdk binary (#206)
e9be6a2 Fix invalid olm.skipRelease with olm.skipRange (#204) (#205)
a1aba29 Allow better controls when generating bundles (#202) (#203)
bf87e35 Drop publisher label for ceilometer (#199)
e943f37 Add support for smartgateway.infra.watch/v2 (#198)
92f9e4a Handle Gnocchi enablement in infrared-openstack.sh (#191)
b4b1e79 Drop deprecated infra.watch/v1alpha1 interface (#197)
fc840cb Use generate_bundle.sh to create the CSV for stf-run-ci role (#195)
91f34d1 move smoketest images away from dockerhub (#194)
a57b4dc Make generating CSVs possible via templated CSV (#193)
25fc522 Update infrared-openstack.sh 13 script (#155)
a0cb2a2 Only need 15 OSDs in test configuration (#187)
600c3fc Install Ansible for working CI (#190)
218d308 (origin/hackfest-working, hackfest-working) Expose SmartGateway debug option (#185)
6514349 Send CPU utilization as a percentage (#184)
4b3e4f4 (fixup/drop_v1alpha1_crd) OSP 16.1 Test Script Updates (#179)
15dd62a Update CI for current STF (master) (#183)
363089d Align to closer upstream base image (#180)
43d6c98 Prep environment for next release (#178)
499ac73 Refactor sg-core (#165)
8bd398e Bump ElasticSearch version to 7.10.2 (#173)
a7b5e2d Resolve issue with ElasticSearch password retrieval (#170)
e3fd69a Use new unstable channel for quickstart (#169)
35791ef Provide support for local building prometheus-webhook-snmp (#166)
bf321ff Set CSV to support downstream release bundle checks (#167)

v1.2.1

Service Telemetry Operator v1.2.1

Minor hotfix changes since the v1.2.0 release.

Changes since v1.2.0

3b66b8f Bump ElasticSearch version to 7.10.2 (#174)
aea6eab Resolve issue with ElasticSearch password retrieval (#170) (#172)
b27417d Set CSV to support downstream release bundle checks (#167) (#168)

v1.2.0

Service Telemetry Operator v1.2.0

The v1.2.0 release of the Service Telemetry Operator builds on the v1.1.0 release with additional functionality for persistent storage configuration, dashboards data sources, and CSV options.

Changes since v1.1.0

c9b9f17 Add Prometheus retention time period control (#163)
590e1a9 Add web interface control for AMQ Interconnect (#162)
9bb5c2d Use new unstable channel for HEAD branch (#161)
6f03676 Set master branch to latest OLM channel (#160)
d853c3c Use hostname -s to create unique cluster hostnames (#153)
f6cbeba [CI] Migrate to GH Actions (#158)
f69935d Make STF deployment optional in OSP13 test harness (#157)
5b7be08 Merge pull request #156 from infrawatch/lmadsen-bump-latest-13
2c6b2df Drop PREFIX and bump OSP13 test scripts to latest puddle
9c24a0e Merge pull request #154 from infrawatch/mrunge-gnocchi-typo
4e41404 Fix typo in parameter name
0bf40f1 Enable libpodstats and sensubility amqp1 transport (#149)
1dde909 Fix smoketest (#152)
0ea38ab Elasticsearch datasources for cloud dashboard (#151)
124b0a5 Merge pull request #150 from infrawatch/csibbitt-1176-fix-storageclass-selection
b37660c Reverting accidental stable channel inclusion
c330b70 Conditional inclusion of storage params
94ad965 Backing out 20Gi -> 20G change
0ed3ec1 Bundle update
a4beb6a Making persistent storage config work

v1.1.0

Service Telemetry Operator Release v1.1.0

The v1.1.0 release of the Service Telemetry Operator (the Operator) is a major re-factor of the CustomResourceDefinition (CRD) from v1alpha1 to v1beta1. Users of the Operator are encouraged to migrate to the new CRD interface. Best effort has been provided to support the legacy CRD interface.

Service Telemetry Operator v1.1.0 is now deployed via a CatalogSource and uses the new bundle method of packaging, which is only compatible with OKD v4.5.

The v1beta1 CRD provides an interface for multi-cloud support to be natively managed by the Operator which has some important features:

• a list of Smart Gateways can be configured via the clouds object instead of the previous method of managing the SmartGateway objects manually outside of the Operator
• Smart Gateways managing events for writing to ElasticSearch via the Elastic Cloud on Kubernetes (ECK) Operator are connected via basic authentication (in order to support the free-tier ElasticSearch). The Operator can now manage the basicAuth configuration of the SmartGateway objects.
• changing the clouds object can remove SmartGateway objects no longer listed in clouds. This is configurable via the cloudsRemoveOnMissing parameter.
• the scrapeInterval for SmartGateway objects can now be configured via the scrapeInterval parameter. This is a global parameter and not configurable per-cloud.
• The Operator can now manage additional parameters for the Grafana objects: disable signout menu, control of whether a route is exposed, administrator username and password.
• New status condition ephemeralStorageEnabled describing if ephemeral storage is enabled for any of the storage components (such as Prometheus, ElasticSearch, or Alertmanager)

Additional changes to the Operator include:

• Listing ServiceTelemetry objects can now be done through the short-name stf, e.g. oc get stf
• The default ServiceTelemetry object name is now default and no longer stf-default
• The naming scheme of SmartGateway objects changes to <STF Name>-<cloud name>-<collector short name>-<data type>, e.g. default-cloud1-ceil-meter, which represents a Smart Gateway for STF instance default, cloud cloud1, collector ceil for ceilometer (and coll for collectd), data type meter for metrics (event for events)
• delivery of SNMP traps via Alertmanager webhooks has been implemented

Commit log since v1.0.3:

f84753a (HEAD -> master, origin/master, origin/HEAD) Provide ability to not deploy stf-connectors.yaml (#146)
643a53f Adjust smoketests and deployment for SNMPtraps (#145)
0b16dc0 Add osp 16.1 deploy with sending metrics (#142)
c064e4f (tag: v1.1.0-rc3) configure storage for ES correctly
f9005ab Set proper value when quickstart in ephemeral mode (#143)
67a970e [master merged] Add snmp traps container and config (#137)
4cbeac2 Add command to enable local registry on CRC
23007c9 Update lastTransitionTime only when changed for ephemeralStorageEnabled status object (#141)
dd62fba Only create GrafanaDatasource when graphing.enabled
96288df Only enable Grafana when graphing is enabled (#140)
8c440b9 (tag: v1.1.0-rc2) Adjust defaults and enhance UI view (#139)
51738b7 Align sample CR with CRD
0178c9e Default channel should be 'stable'
0589bb7 (crd-v1) Remove components that are marked disabled (#138)
bad46af Changes towards 'latest' deploy in quickstart (#136)
acd3869 lmadsen scorecard cleanup (#135)
644e40a Post-refactor clean up (#134)
88b245b (tag: v1.1.0-rc1) Update infrared-openstack.sh
c34614c Update infrared-openstack.sh
ecd63b5 Dictionary-based API and Multi-Cloud Support (#119)
bd478f0 Add operator-sdk Ansible collection (#133)
71c2a4f Update CI automation (#132)
364c650 Migrate to bundle format (#131)
202019b Add prefix to virtual resources (#130)
aad41e6 Adjust load alerts to account for CPU count (#127)
d33b59d Adjust release deployment scripts (#126)
de0a349 Merge pull request #122 from infrawatch/csibbitt-791-797-ha-sg-am-prom
cf555ae Fix whitespace in sg metrics template conditionals
ea94dbb Merge pull request #124 from infrawatch/mrunge_workload
4cac0d0 Add running a workload at the end of deploy
7228390 Update roles/servicetelemetry/templates/manifest_smartgateway_metrics.j2
617fc90 Update roles/servicetelemetry/templates/manifest_smartgateway_events.j2
f2fb375 HA for SmartGateway, AlertManager, and Prometheus
aeda917 Full ceilometer/collectd metrics/events + UI (#121)
5d0bd47 Use valid parameter for PVC in ES (#118)
bafb195 Update README.md
944ca7f Basic Graphing Support (#115)
62cf657 Add basic Makefile for local builds (#116)
f53c476 (origin/csibbitt-797-HA-Prom-and-alertmanager) Promxy POC in test directory (#114)
9d6684b Bump Operator-SDK base image to v0.16.0 (#113)
55e20b3 Support for Smart Gateway compile changes (#112)
6c48937 Use passed_phase2 for test clusters (OSP16) (#111)

v1.1.0-rc3

Service Telemetry Operator Release Candidate 3

The v1.1.0-rc3 (release candidate) of the Service Telemetry Operator (the Operator) is a bugfix release based on testing of RC2. Additionally, delivery of SNMP traps via Alertmanager webhooks has been implemented.

ChangeLog since v1.1.0-rc2:

c064e4f (HEAD -> master, origin/master, origin/HEAD) configure storage for ES correctly
f9005ab Set proper value when quickstart in ephemeral mode (#143)
67a970e [master merged] Add snmp traps container and config (#137)
4cbeac2 Add command to enable local registry on CRC
23007c9 Update lastTransitionTime only when changed for ephemeralStorageEnabled status object (#141)
dd62fba Only create GrafanaDatasource when graphing.enabled
96288df Only enable Grafana when graphing is enabled (#140)

v1.1.0-rc2

Service Telemetry Operator Release Candidate 2

The v1.1.0-rc2 (release candidate) of the Service Telemetry Operator (the Operator) is a major re-factor of the CustomResourceDefinition (CRD) from v1alpha1 to v1beta1. Users of the Operator are encouraged to migrate to the new CRD interface. Best effort has been provided to support the legacy CRD interface.

This release candidate primarily addresses 2 major focuses:

• getting the bundle manifests in the best possible shape
• providing a better user experience when upgrading the Operator or creating new ServiceTelemetry objects

Bundles are now generated via operator-sdk 0.19.4 vs 0.17.2. This provides better scorecard validation and a simpler interface to building the bundles. Adjustments based on the scorecard recommendations were also applied. A new CSV channel of latest was added for an easier testing environment and previous release of STF components were properly added to the stable channel. The stable channel was marked as the default channel for Subscriptions to the CatalogSource.

The sample CustomResource was updated to provide all options and their defaults when creating a new ServiceTelemetry object in the OpenShift user interface. A large refactoring of the UI decorators (x-descriptors) was done to provide better controls to administrators who wish to deploy from the OpenShift UI rather than the console commands as documented.

The quickstart.sh script was updated fully to allow end-to-end setup and environment smoketests to pass and run through to completion.

The Operator was updated to remove objects that were previously marked enabled: true and subsequently changed to enabled: false.

ChangeLog since v1.1.0-rc1:

8c440b9 (HEAD -> master, origin/master, origin/HEAD) Adjust defaults and enhance UI view (#139)
51738b7 Align sample CR with CRD
0178c9e Default channel should be 'stable'
0589bb7 (crd-v1) Remove components that are marked disabled (#138)
bad46af Changes towards 'latest' deploy in quickstart (#136)
acd3869 lmadsen scorecard cleanup (#135)
644e40a Post-refactor clean up (#134)