* Update istio test ref to 1.20.0-rc.0
* Remove added toolchain line
* Run make gen
* Update to set ENABLE_EXTERNAL_NAME_ALIAS to false (default) for testing
* Move gateway tcp-traffic shifting to demo
* Change gateway tcp-traffic shifting to not remove istio in demo...
* Change gateway tcp-traffic shifting to remove istiooperator in demo...
* Update to latest 1.20 commit
* Change to be a minimal test
Users deploying in Openshift cluster have reported that upgrade from Istio 1.17 to Istio 1.18.3 has failed due to the lack of kernel module xt_multiport from the cluster.
This is most probably related to modifications in IPtables creation, see 7e7d52a2d1/tools/istio-iptables/pkg/capture/run.go (L369)
Signed-off-by: Faseela K <faseela.k@est.tech>
* Update Istio VM installation guide in multi-network setup
Add missing instruction on Istio VM installation guide in multi-network setup
* Update Istio VM installation guide in multi-network setup
* updating primary-remote on multi-net docs
* adding missing snips of make gen
* updating primary-remote_multi-network test.sh
* Update content/en/docs/setup/install/multicluster/primary-remote_multi-network/index.md
Co-authored-by: Frank Budinsky <frankb@ca.ibm.com>
* Update content/en/docs/setup/install/multicluster/primary-remote_multi-network/index.md
Co-authored-by: Frank Budinsky <frankb@ca.ibm.com>
* adjusting the sequence of test func calls
---------
Co-authored-by: Frank Budinsky <frankb@ca.ibm.com>
* Update istio test ref, do not strip sha from gateway version
* Rework sha, remove code to update gateway-api automatically
* Update Makefile to use full sha
* Some minor tweaks
* Some minor tweaks
* Move sha logic to a bash script
* Remove extraneous expected line in dns-proxy test
* fix tab/space
* adding Amazon EKS with Istio setup to platform-setup page
* adding Amazon EKS with Istio setup to platform-setup page
* updating the test section of the doc
* adding period at the end of description for the error in test
* adding newline at end of file
* changing the weight to make it ordered alphabetically.
* updating the link as it was broken due to this change made to folder structure in upstream https://github.com/aws-ia/terraform-aws-eks-blueprints/pull/1754
* adding Amazon EKS with Istio setup to platform-setup page
* adding Amazon EKS with Istio setup to platform-setup page
* updating the test section of the doc
* adding period at the end of description for the error in test
* adding newline at end of file
* changing the weight to make it ordered alphabetically.
* Remove anyuid instructions for OpenShift
These instructions are no longer needed after these changes https://github.com/istio/istio/pull/45394
* Removed blank line that caused failure
* Update to latest istio commit. Also use rc for gateway-api.
* Fix Makefile update-gateway-version
* Update istio to 1.19.0-rc1
* Remove wasm from envoy test
* Update gateway-api to rc-2.
* Revert gateway-api t0 0.7.1
* Update samples for core group
* Revert gateway-api to rc.1 to match istio repo
The regex was @.*@ which will match anything between two @'s, including a command like:
`k3d cluster create --api-port 6550 -p '9080:80@loadbalancer' -p '9443:443@loadbalancer' --agents 2 --k3s-arg '--disable=traefik@server:*'`
The regex should be something that matches valid URL syntaxes.
Signed-off-by: Daniel Hawton <daniel.hawton@solo.io>
* Re-enable processing of gateway-api version in `make gen`
* Update gateway-api to be the publicly released base version
* remove script and add awk in Makefile
* rebase and make gen
* remove injected annotation for istio-cni
* add injector webhook config back, add tip about it overwriting existing configs
* update docs with istio_cni.chained installation config
* update helm install instructions
* fix linting
* pass linter
* clarify tip for previous Openshift users to be aware that the injection webhook is not required
* update per PR suggestion
* generate new snips
* update cni docs by removing tip and adding openshift install command per PR
* lint fix
---------
Co-authored-by: Jacob Delgado <jacob.delgado@volunteers.acasi.info>
* Change Istio Classic terminology to Istio APIs
Signed-off-by: Keith Mattix II <keithmattix@microsoft.com>
* Fix stutter
Signed-off-by: Keith Mattix II <keithmattix@microsoft.com>
* Change back parallel structure formation
Signed-off-by: Keith Mattix II <keithmattix@microsoft.com>
---------
Signed-off-by: Keith Mattix II <keithmattix@microsoft.com>
* Master has a non release gateway API. Changes to allow that.
* Grab a istio test ref from main branch
* <ove from istio.io/pkg to istio/istio/pkg
* revert one change
* update istio/api
* Run make gen
* disable failing test temporarily
* generic-args-md
* create generic boilerplates/args.md
* include gw_api_version, full_version, previous_version
* update install/operator/test.sh to source and use new vars
* update test/util/gateway-api.sh to source and use new gw var
* remove gateway-api-version.md
* relates to #13085
* fix linting fail for args.md
* remove sed command for revision string, use param expansion
* Enable tests for operator install
* install-operator-tests
* update snip_id
* set snip_id=none for non-tested snips
* add test scenarios for helm,watchns,install istio,update
* use functions for better readability
* add newline at the end
* install-operator-tests
* add tests for inplace and canary upgrade
* update version 1.8.1 and 1.9.0 to 1.17.1 and 1.17.2 in docs
* disable snip remove_operator failing lint/test
* install-operator-tests
* remove all hardcoded versions in operator install doc
* add istio_previous_version and full/previous revision to snip.py
* adjust test.sh accordingly
* install-operator-tests
* add shorcodes for revision versions
* install-operator-tests
* fix canary upgrade (upgrade from previous to full version)
* install-operator-tests
* fix reference to full_version in canary upgrade
* install-operator-tests
* fix slightly off format
* install-operator-tests
* disable test for upgrade with helm tip
* add instructions to doc for canary upgrade prep
* update test.sh and snips.sh accordingly
* install-operator-tests
* fix profile on upgrade, use default
Before the Deployment was kept as minimal as possible. This is mostly
good, because users can set any Deployment settings and things work.
However, running as non-root is a bit quirky, and very very strongly
recommended, so I think its worth calling out here.
This is one of the #1 sources of confusion we get.