* Replace all cluster roles with autogenerated namespaced roles
* Use cluster role only for preupgrade api extension permissions
* Correct role definition typos
Signed-off-by: Sanket Sudake <sanketsudake@gmail.com>
Co-authored-by: shaunak_deshmukh <shaunak@infracloud.io>
Co-authored-by: Sanket Sudake <sanketsudake@gmail.com>
* Convert ClusterRoles to Roles for all components for multiple namespaces
* Added templates for rules and role generator to dynamically generate roles for each component
* Fix role-generator component templating
* Enable namespaces variable for preupgradechecks
Signed-off-by: Sanket Sudake <sanketsudake@gmail.com>
Co-authored-by: shaunak_deshmukh <shaunak@infracloud.io>
Co-authored-by: Sanket Sudake <sanketsudake@gmail.com>
* Separate service accounts for each component
* Permission changes per component
* Fixed fluentbit permissions
* added hooks to pre-upgrade service account and role
* Adjusted hook weights for pre-upgrade hook
* Replaced * with explicit permissions
Following component has been updated with securityContext -
1. buildermgr
2. controller
3. executor
4. kubewatcher
5. router
6. timer
7. storagesvc
* Removed the security context from fetcher as per the offline discussion
* fix helm templating on resources of router deployment
* add resources attribute to other fission deployments
Co-authored-by: crmendes <carlos.mendes@kelvininc.com>
Currently default pod termination logs go to /dev/termination-log.
With CI we need to change customize path somewhere to /var/log for
exporting logs with the kind export logs command.
Setting FallbackToLogsOnError as termination policy for skaffold.
Signed-off-by: Sanket Sudake <sanketsudake@gmail.com>
In Fission CI as well in a couple of scenarios, noticed Fission pods
getting preempted when Function pods are getting created. As both
Fission components and function pods have the same priority class
fission component pods might get preempted. It to be recommended to
set higher priority class for components such as Fission router, executor,
controller so that Fission functioning doesn't get disrupted.
Most of the Fission components work fine with restart except the executor. So
user should set priorityClass for Fission executor.
Signed-off-by: Sanket Sudake <sanketsudake@gmail.com>
* Sync a few files from fission-all to fission-core chart
* Change values in fission-all chart
We keep the following components by default disabled now, which were enabled by default earlier.
1. nats - Fission Nats integration
2. influxdb - Influxdb and logger component
3. prometheus - Prometheus disabled by default
4. canaryDeployment - Disabled by default
This change reduces the need for a fission-core chart and we can configure
values of fission-all so that it can provide the behaviour of fission-all
as well as fission-core.
* Remove fission-core chart
* Add README in the fission-all chart
* Rename clusterrolebinding fission-crd to fission-cr-admin
* Add icon and sources in helm chart
Signed-off-by: Sanket Sudake <sanketsudake@gmail.com>
