* bump prometheus to the latest v2.19.3
latest prometheus version shows a lot of decrease in the memory usage
and other benefits
Signed-off-by: Tarun Pothulapati <tarunpothulapati@outlook.com>
* CNI add support for priorityClassName
As requested in #2981 one should be able to optionally define a priorityClassName for the linkerd2 pods.
With this commit support for priorityClassName is added to the CNI plugin helm chart as well as to the
cli command for installing the CNI plugin.
Also added an `installNamespace` Helm option for the CNI installation.
Implements part of #2981.
Signed-off-by: alex.berger@nexiot.ch <alex.berger@nexiot.ch>
This pr adds `globa.prometheusUrl` field which will be used to configure publlic-api, hearbeat, grafana, etc (i,e query path) to use a external Prometheus.
* support overriding inbound and outbound connect timeouts.
* add validation on user provided TCP connect timeouts
* convert valid time values into ms
Signed-off-by: Matt Miller <mamiller@rosettastone.com>
* Add sidecar container support for linkerd-prometheus
Adds a new setting to the Prometheus' Helm config, allowing adding any kind of sidecar containers to the main container.
The specific use case that inspired this was for exporting data from Prometheus to external systems (e.g. cloudwatch, stackdriver, datadog) using a process that watches the prometheus write-ahead log (WAL).
Signed-off-by: Nathan J. Mehl <n@oden.io>
Add a new structure on the destination controller side to keep track of contextual information.
The token format has been changed from ns:<namespace> to a JSON format so that more variables can be
encdoed in the token. As part of this PR, a new field 'nodeName' has been added to help with service
topologies.
Fixes#4498
Signed-off-by: Matei David <matei.david.35@gmail.com>
This PR removes the service mirror controller from `linkerd mc install` to `linkerd mc link`, as described in https://github.com/linkerd/rfc/pull/31. For fuller context, please see that RFC.
Basic multicluster functionality works here including:
* `linkerd mc install` installs the Link CRD but not any service mirror controllers
* `linkerd mc link` creates a Link resource and installs a service mirror controller which uses that Link
* The service mirror controller creates and manages mirror services, a gateway mirror, and their endpoints.
* The `linkerd mc gateways` command lists all linked target clusters, their liveliness, and probe latences.
* The `linkerd check` multicluster checks have been updated for the new architecture. Several checks have been rendered obsolete by the new architecture and have been removed.
The following are known issues requiring further work:
* the service mirror controller uses the existing `mirror.linkerd.io/gateway-name` and `mirror.linkerd.io/gateway-ns` annotations to select which services to mirror. it does not yet support configuring a label selector.
* an unlink command is needed for removing multicluster links: see https://github.com/linkerd/linkerd2/issues/4707
* an mc uninstall command is needed for uninstalling the multicluster addon: see https://github.com/linkerd/linkerd2/issues/4708
Signed-off-by: Alex Leong <alex@buoyant.io>
* Migrate CI to docker buildx and other improvements
## Motivation
- Improve build times in forks. Specially when rerunning builds because of some flaky test.
- Start using `docker buildx` to pave the way for multiplatform builds.
## Performance improvements
These timings were taken for the `kind_integration.yml` workflow when we merged and rerun the lodash bump PR (#4762)
Before these improvements:
- when merging: `24:18`
- when rerunning after merge (docker cache warm): `19:00`
- when running the same changes in a fork (no docker cache): `32:15`
After these improvements:
- when merging: `25:38`
- when rerunning after merge (docker cache warm): `19:25`
- when running the same changes in a fork (docker cache warm): `19:25`
As explained below, non-forks and forks now use the same cache, so the important take is that forks will always start with a warm cache and we'll no longer see long build times like the `32:15` above.
The downside is a slight increase in the build times for non-forks (up to a little more than a minute, depending on the case).
## Build containers in parallel
The `docker_build` job in the `kind_integration.yml`, `cloud_integration.yml` and `release.yml` workflows relied on running `bin/docker-build` which builds all the containers in sequence. Now each container is built in parallel using a matrix strategy.
## New caching strategy
CI now uses `docker buildx` for building the container images, which allows using an external cache source for builds, a location in the filesystem in this case. That location gets cached using actions/cache, using the key `{{ runner.os }}-buildx-${{ matrix.target }}-${{ env.TAG }}` and the restore key `${{ runner.os }}-buildx-${{ matrix.target }}-`.
For example when building the `web` container, its image and all the intermediary layers get cached under the key `Linux-buildx-web-git-abc0123`. When that has been cached in the `main` branch, that cache will be available to all the child branches, including forks. If a new branch in a fork asks for a key like `Linux-buildx-web-git-def456`, the key won't be found during the first CI run, but the system falls back to the key `Linux-buildx-web-git-abc0123` from `main` and so the build will start with a warm cache (more info about how keys are matched in the [actions/cache docs](https://docs.github.com/en/actions/configuring-and-managing-workflows/caching-dependencies-to-speed-up-workflows#matching-a-cache-key)).
## Packet host no longer needed
To benefit from the warm caches both in non-forks and forks like just explained, we're required to ditch doing the builds in Packet and now everything runs in the github runners VMs.
As a result there's no longer separate logic for non-forks and forks in the workflow files; `kind_integration.yml` was greatly simplified but `cloud_integration.yml` and `release.yml` got a little bigger in order to use the actions artifacts as a repository for the images built. This bloat will be fixed when support for [composite actions](https://github.com/actions/runner/blob/users/ethanchewy/compositeADR/docs/adrs/0549-composite-run-steps.md) lands in github.
## Local builds
You still are able to run `bin/docker-build` or any of the `docker-build.*` scripts. And to make use of buildx, run those same scripts after having set the env var `DOCKER_BUILDKIT=1`. Using buildx supposes you have installed it, as instructed [here](https://github.com/docker/buildx).
## Other
- A new script `bin/docker-cache-prune` is used to remove unused images from the cache. Without that the cache grows constantly and we can rapidly hit the 5GB limit (when the limit is attained the oldest entries get evicted).
- The `go-deps` dockerfile base image was changed from `golang:1.14.2` (ubuntu based) to `golang-1:14.2-alpine` also to conserve cache space.
# Addressed separately in #4875:
Got rid of the `go-deps` image and instead added something similar on top of all the Dockerfiles dealing with `go`, as a first stage for those Dockerfiles. That continues to serve as a way to pre-populate go's build cache, which speeds up the builds in the subsequent stages. That build should in theory be rebuilt automatically only when `go.mod` or `go.sum` change, and now we don't require running `bin/update-go-deps-shas`. That script was removed along with all the logic elsewhere that used it, including the `go_dependencies` job in the `static_checks.yml` github workflow.
The list of modules preinstalled was moved from `Dockerfile-go-deps` to a new script `bin/install-deps`. I couldn't find a way to generate that list dynamically, so whenever a slow-to-compile dependency is found, we have to make sure it's included in that list.
Although this simplifies the dev workflow, note that the real motivation behind this was a limitation in buildx's `docker-container` driver that forbids us from depending on images that haven't been pushed to a registry, so we have to resort to building the dependencies as a first stage in the Dockerfiles.
EndpointSlices have been made opt-in due to their experimental nature. This PR
introduces a new install flag 'enableEndpointSlices' that will allow adopters to
specify in their cli install or helm install step whether they would like to
use endpointslices as a resource in the destination service, instead of the
endpoints k8s resource.
Signed-off-by: Matei David <matei.david.35@gmail.com>
This moves Prometheus as a add-on, thus making it optional but enabled by default. The also make `linkerd-prometheus` more configurable, and allow it to have its own life-cycle for upgrades, configuration, etc.
This work will be followed by documentation that help users configure existing Prometheus to work with Linkerd.
**Changes Include:**
- moving prometheus manifests into a separate chart at `charts/add-ons/prometheus`, and adding it as a dependency to `linkerd2`
- implement the `addOn` interface to support the same with CLI.
- include configuration in `linkerd-config-addons`
**User Facing Changes:**
The default install experience does not change much but for users who have already configured Prometheus differently, would need to apply the same using the new configuration fields present in chart README
The splitStringListToPorts helm function is currently incorrectly formating a list of ports as an array of Port objects that look ike {"port" : 555}. The config map protobuf representation however expects that the ignoreOutboundPorts and ignoreInboundPorts fields are are list of PortRange objects ({"portRange" : 555}).
This was causing the injector to return an empty string when trying to parse a PortRange object resulting in the ports not getting set correctly when injecting workloads. Note that this is happening only with helm installations as this is when we are actually using a helm template for outputting the config map.
To fix that the splitStringListToPorts helm function is changed to format the objects as the json representation of PortRange and is renamed to splitStringListToPortRanges
Fix: #4679
Signed-off-by: Zahari Dichev zaharidichev@gmail.com
* update helm render tests to read child charts values.yaml
Helm installation by default, considers values.yaml for dependend charts
and uses them in rendering. This function is being used for add-ons to
keep the default template values, allowing further overriden from the
parent chart's i.e linkerd2 values.yaml or --addon-config through CLI.
This PR updates the Helm tests to reflect the same i.e consider
values.yaml of chart dependencies if present.
This does not have any UX changes but helps with the follow up
add-on related work.
Using following command the wrong spelling were found and later on
fixed:
```
codespell --skip CHANGES.md,.git,go.sum,\
controller/cmd/service-mirror/events_formatting.go,\
controller/cmd/service-mirror/cluster_watcher_test_util.go,\
SECURITY_AUDIT.pdf,.gcp.json.enc,web/app/img/favicon.png \
--ignore-words-list=aks,uint,ans,files\' --check-filenames \
--check-hidden
```
Signed-off-by: Suraj Deshmukh <surajd.service@gmail.com>
Based on the [EndpointSlice PR](https://github.com/linkerd/linkerd2/pull/4663), this is just the k8s/api support for endpointslices to shorten the first PR.
* Adds CRD
* Adds functions that check whether the cluster has EndpointSlice access
* Adds discovery & endpointslice informers to api.
Signed-off-by: Matei David <matei.david.35@gmail.com>
* feat: add log format annotation and helm value
Json log formatting has been added via https://github.com/linkerd/linkerd2-proxy/pull/500
but wiring the option through as an annotation/helm value is still
necessary.
This PR adds the annotation and helm value to configure log format.
Closes#2491
Signed-off-by: Naseem <naseem@transit.app>
Data disappears upon prometheus restarts due to it being all in-memory.
Adding an option to enabled persistence by means of a PVC would be the right approach. It is commonly seen in a wide array of helm charts.
Fixes#4576
Signed-off-by: Naseem <naseem@transit.app>
Regenerated protobuf files, using version 1.4.2 that was upgraded from
1.3.2 with the proxy-api update in #4614.
As of v1.4 protobuf messages are disallowed to be copied (because they
hold a mutex), so whenever a message is passed to or returned from a
function we need to use a pointer.
This affects _mostly_ test files.
This is required to unblock #4620 which is adding a field to the config
protobuf.
* Update inject to error out on failure
Update injection process to throw an error when the reason for failure is due to sidecar, udp, automountServiceAccountToken or hostNetwork
Signed-off-by: Mayank Shah <mayankshah1614@gmail.com>
In #4585 we are observing an issue where a loop is encountered when using nginx ingress. The problem is that the outbound proxy does a dst lookup on the IP address which happens to be the very same address the ingress is listening on.
In order to avoid situations like that this PR introduces a way to modify the set of networks for which the proxy shall do IP based discovery. The change introduces a helm flag `.Values.global.proxy.destinationGetNetworks` that can be used to modify this value. There are two ways a user can affect the this setting:
- setting the `destinationGetNetworks` field in values during a Helm install, which changes the default on all injected pods
- using an annotation ` config.linkerd.io/proxy-destination-get-networks` for injected workloads to override this value
Note that this setting cannot be tweaked through the `install` or `inject` command
Fix: #4585
Signed-off-by: Zahari Dichev <zaharidichev@gmail.com>
Fixes#4606
This has not worked as far back as stable-2.6.0.
## Solution
The recommended upgrade process is to include `--prune` as part of `kubectl
apply ..`:
```bash
$ linkerd upgrade | kubectl apply --prune -l linkerd.io/control-plane-ns=linkerd -f -
```
This is an issue for multi-stage upgrade because `linkerd upgrade config` does
not include the `linkerd-config` ConfigMap in it's output.
`kubectl apply --prune ..` will then prune this resource because it matches the
label selector *and* is not in the above output.
The issue occurs when `linkerd upgrade control-plane` is run and expects to find
the ConfigMap that was just pruned.
This can be fixed by not suggesting to prune resources as part of the
multi-stage upgrade.
## Considered
Including `templates/config.yaml` in the install output regardless of the stage.
Instead of it being a template only used in `control-plane` stage in
[render](4aa3ca7f87/cli/cmd/install.go (L873-L886)), it could always be rendered.
This just exposes other things that are pruned in the process:
```bash
❯ bin/linkerd upgrade control-plane |kubectl apply --prune -l linkerd.io/control-plane-ns=linkerd -f -
× Failed to build upgrade configuration: secrets "linkerd-identity-issuer" not found
For troubleshooting help, visit: https://linkerd.io/upgrade/#troubleshooting
error: no objects passed to apply
```
Ultimately, resources part of the `control-plane` stage need to remain and that
will not happen if we prune all resources not in the `config` stage output
Signed-off-by: Kevin Leimkuhler <kevin@kleimkuhler.com>
As reported in #4259 linkerd check run from linkerd's web cconsole is
broken as the underlying RBAC Role cannot access the apiregistration.k8s.io API Group.
With this commit the RBAC Role is fixed allowing read-only access to the API Group
apiregistration.k8s.io.
Fixes#4259
Signed-off-by: alex.berger@nexiot.ch <alex.berger@nexiot.ch>
Put back space after `grafanaUrl` label in `linkerd-config-addons.yaml`
to avoid breaking the yaml parsing.
```
$ linkerd check
...
linkerd-addons
--------------
‼ 'linkerd-config-addons' config map exists
could not unmarshal linkerd-config-addons config-map: error
unmarshaling JSON: while decoding JSON: json: cannot unmarshal
string into Go struct field Values.global of type linkerd2.Global
```
This was added in #4544 to avoid having the configmap being badly formatted.
So this PR fixes the yaml, but then if we don't set `grafanaUrl` the
configmap format gets messed up, but apparently that's just a cosmetic
problem:
```
apiVersion: v1
data:
values: "global:\n grafanaUrl: \ngrafana:\n enabled: true\n
image:\n name:
gcr.io/linkerd-io/grafana\n name: linkerd-grafana\n resources:\n
cpu:\n limit:
240m\n memory:\n limit: null\ntracing:\n enabled:
false"
kind: ConfigMap
```
Fixes#4541
This PR adds the following checks
- if a mirrored service has endpoints. (This includes gateway mirrors too).
- if an exported service is referencing a gateway that does not exist.
Signed-off-by: Tarun Pothulapati <tarunpothulapati@outlook.com>
Signed-off-by: Alex Leong <alex@buoyant.io>
Co-authored-by: Alex Leong <alex@buoyant.io>
* Add namespace global flag to hold default namespace name (#4469)
Signed-off-by: Matei David <matei.david.35@gmail.com>
* Change name of controlplane install namespace constant and init point for kubeNamespace
Signed-off-by: Matei David <matei.david.35@gmail.com>
Problem
When updating / writing tests with complex data, e.g the certificates, the build-in diff is not as powerful as dedicated external tool.
Solution
Dump all resource specifications created as part of failing tests to a supplied folder for external analysis.
Signed-off-by: Lutz Behnke <lutz.behnke@finleap.com>
Fixes#4454
As explained
[here](https://github.com/kubernetes/kubernetes/issues/36222#issuecomment-553966166),
trailing spaces in configmap data makes it to look funky when retrieved
later on. This is currently affecting `linkerd-config-addons` and
`linkerd-gateway-config`:
```
$ k -n linkerd-multicluster get cm linkerd-gateway-config -oyaml
apiVersion: v1
data:
nginx.conf: "events {\n}\nstream { \n
\ server { \n
\ listen 4180; \n
\ proxy_pass 127.0.0.1:4140; \n
\ } \n}
\nhttp {\n server {\n listen 4181;\n location /health {\n access_log
off;\n return 200 \"healthy\\n\";\n }\n }\n server {\n listen
\ 8888;\n location /health-local {\n access_log off;\n return
200 \"healthy\\n\";\n }\n } \n}"
kind: ConfigMap
```
AFAIK this is only cosmetic and doesn't affect functionality.
* Fixes#4305
Fixed SP route for `POST /api/v1/query`:
```
$ bin/linkerd routes -n linkerd deploy/linkerd-prometheus
ROUTE SERVICE SUCCESS RPS LATENCY_P50 LATENCY_P95 LATENCY_P99
GET /api/v1/query_range linkerd-prometheus 100.00% 3.9rps 1ms 2ms 2ms
GET /api/v1/series linkerd-prometheus 100.00% 1.1rps 1ms 1ms 1ms
POST /api/v1/query linkerd-prometheus 100.00% 3.1rps 1ms 17ms 19ms
[DEFAULT] linkerd-prometheus - - - - -
```
Also added one missing route for `linkerd-grafana`, realizing afterwards there are
many other ones missing, but not really worth adding them all.
I also removed the routes in `linkerd-controller` for the tap routes
given that's no longer handled in that service.
And the tap service SP was also removed alltogether since nothing was
getting reported.
Change terminology from local/remote to source/target in `multicluster` CLI help
text.
This does not change any variable, function, struct, or field names since
testing is still improving.
Relevant issue: #4480
Signed-off-by: Kevin Leimkuhler <kevin@kleimkuhler.com>
There are a few notable things happening in this PR:
- the probe manager has been decoupled from the cluster_watcher. Now its only responsibility is to watch for mirrored gateways beeing created and to probe them. This means that probes are initiated for all gateways no matter whether there are mirrored services being paired
- the number of paired services is derived from the existing services in the cluster rather than being published as a metric by the prober
- there are no events being exchanged between the cluster watcher and the probe manager
Signed-off-by: Zahari Dichev <zaharidichev@gmail.com>
Failures in `bin/_test-run` from commands different than `go test`
aren't currently properly reported, in part because CI's bash default is
to have `set -e` which terminates the script and just outputs
`##[error]Process completed with exit code 2.` like
[here](https://github.com/linkerd/linkerd2/pull/4496/checks?check_run_id=720720352#step:14:116)
```
linkerd-existence
-----------------
√ 'linkerd-config' config map exists
√ heartbeat ServiceAccount exist
√ control plane replica sets are ready
× no unschedulable pods
linkerd-controller-6c77c7ffb8-w8wh5: 0/1 nodes are available: 1 node(s) had taint {node.kubernetes.io/not-ready: }, that the pod didn't tolerate.
linkerd-destination-6767d88f7f-rcnbq: 0/1 nodes are available: 1 node(s) had taint {node.kubernetes.io/not-ready: }, that the pod didn't tolerate.
linkerd-grafana-76c76fcfb9-pdhfb: 0/1 nodes are available: 1 node(s) had taint {node.kubernetes.io/not-ready: }, that the pod didn't tolerate.
linkerd-identity-5bcf97d6c8-q6rll: 0/1 nodes are available: 1 node(s) had taint {node.kubernetes.io/not-ready: }, that the pod didn't tolerate.
linkerd-prometheus-6b95c56b44-hd9m6: 0/1 nodes are available: 1 node(s) had taint {node.kubernetes.io/not-ready: }, that the pod didn't tolerate.
linkerd-proxy-injector-58d794ff9-jf7cj: 0/1 nodes are available: 1 node(s) had taint {node.kubernetes.io/not-ready: }, that the pod didn't tolerate.
linkerd-sp-validator-6c5f999bfb-qg252: 0/1 nodes are available: 1 node(s) had taint {node.kubernetes.io/not-ready: }, that the pod didn't tolerate.
linkerd-tap-6fdf84fc65-6txvr: 0/1 nodes are available: 1 node(s) had taint {node.kubernetes.io/not-ready: }, that the pod didn't tolerate.
linkerd-web-8484fbd867-nm8z2: 0/1 nodes are available: 1 node(s) had taint {node.kubernetes.io/not-ready: }, that the pod didn't tolerate.
see https://linkerd.io/checks/#l5d-existence-unschedulable-pods for hints
Status check results are ×
[error]Process completed with exit code 2.
```
I've made the following changes to `bin/_test-run` to generate better
messages and Github annotations when an error occurs:
- Unset `set -e` so that errors don't immediately exit the script and
don't allow us to properly format the errors.
- Removed many of the `exit_on_err` calls after go test calls because
those output enough information already (they were not being used
anyways in CI because of `set -e`). And instead have `run_test` exit
upon a `go test` error.
- Added `exit_on_err` calls right after non-`go-test` commands to
properly report their failure.
- Refactored the `exit_on_err` function so that it generates a Github
error annotation upon failure.
- Removed `trap` in `install_stable`, since the OS should be able to
handle GC for stuff under `/tmp`.
Also, I've changed the exit 2 code from `linkerd check` when it fails,
to exit code 1.
Quoting the list of directories passed to `goimports` was causing the list to be interpreted as a single argument which was stopping `bin/fmt` from working.
Instead, use `read` to split the list of directories into an array.
Also fix up incorrect formatting that has crept in while `bin/fmt` has been broken.
Signed-off-by: Alex Leong <alex@buoyant.io>
This change adds a `allow` and `link` commands, effectivelly enabling a cluster to have more than one set of credentials that allow it to be mirrored.
Fx #4461
Signed-off-by: Zahari Dichev <zaharidichev@gmail.com>
Co-authored-by: Alex Leong <alex@buoyant.io>
Rajat Jindal
cpretzer
Alex Leong
Tarun Pothulapati
Alexander Berger
Naseem
Matt Miller
memory
Matei David
Alejandro Pedraza
Wei Lun
Zahari Dichev
Desmond Ho
Suraj Deshmukh
Arthur Silva Sens
Oliver Gould
Mayank Shah
Lutz Behnke
Kevin Leimkuhler
Joakim Roubert
Lutz Behnke