fixes: CCI-236
flag = none
Test-Plan:
get it passing as is, then make a version of this
PS have a bunch of test changes to force the distributed
path to trigger.
- test with one node
-- PS 11
-- https://jenkins.inst-ci.net/job/Canvas/job/test-suites/job/flakey-spec-catcher/299/
- test with multiple nodes
-- PS 14
-- https://jenkins.inst-ci.net/job/Canvas/job/test-suites/job/flakey-spec-catcher/306/
and for testing with no spec changes, the latest PS will have no tests
Change-Id: I4c69ce5c4cb67448ed270ae9fb3255ce21356b11
Reviewed-on: https://gerrit.instructure.com/c/canvas-lms/+/227214
Tested-by: Service Cloud Jenkins <svc.cloudjenkins@instructure.com>
Reviewed-by: James Butters <jbutters@instructure.com>
Reviewed-by: Jacob Powell <spowell@instructure.com>
QA-Review: Jacob Powell <spowell@instructure.com>
Product-Review: Rex Fleischer <rfleischer@instructure.com>
flag = none
Change-Id: Ice5002596684c5000d1b8190b75dd16e7cf04a6f
Reviewed-on: https://gerrit.instructure.com/c/canvas-lms/+/227057
Reviewed-by: Jacob Powell <spowell@instructure.com>
Reviewed-by: Rex Fleischer <rfleischer@instructure.com>
Tested-by: Service Cloud Jenkins <svc.cloudjenkins@instructure.com>
QA-Review: James Butters <jbutters@instructure.com>
Product-Review: James Butters <jbutters@instructure.com>
Decrease the timeout to a more acceptable time. Don't run FSC
on change-merged events.
flag = none
Change-Id: I4e2a4ceae64462f886c194e9c79977027b90c7f7
Reviewed-on: https://gerrit.instructure.com/c/canvas-lms/+/226771
Tested-by: Service Cloud Jenkins <svc.cloudjenkins@instructure.com>
Reviewed-by: Jacob Powell <spowell@instructure.com>
QA-Review: James Butters <jbutters@instructure.com>
Product-Review: James Butters <jbutters@instructure.com>
Add stage to copy over output from fsc for debugging. Add
cleanup post steps.
flag = none
Change-Id: Ibe8795e114f23efea86488285fe788bc8347a8de
Reviewed-on: https://gerrit.instructure.com/c/canvas-lms/+/226416
Tested-by: Service Cloud Jenkins <svc.cloudjenkins@instructure.com>
Reviewed-by: Jacob Powell <spowell@instructure.com>
QA-Review: James Butters <jbutters@instructure.com>
Product-Review: James Butters <jbutters@instructure.com>
To optimize re-runs, FSC uses the same settings for
RSpec::Core::Runner but just clears the RSpec example queue
between runs. The changes in spec/selenium/common.rb ensure
that we don't start up a new webserver and driver when there
is already one existing. This creates connection issues.
closes: CCI-225
Test-plan:
> export COMPOSE_FILE='docker-compose.new-jenkins.yml:docker-compose.new-jenkins-flakey-spec-catcher.yml:docker-compose.new-jenkins-selenium.yml'
> docker-compose up -d
> docker-compose run --rm web flakey_spec_catcher
- verify that no specs are re-run
- Try making changes to other selenium and api specs, re-run
> docker-compose run --rm web flakey_spec_catcher
- and make sure that the results appear as expected
Change-Id: I005ca9dec5e75f72788a7db5a9fa640f69071f6d
Reviewed-on: https://gerrit.instructure.com/c/canvas-lms/+/218480
Tested-by: Service Cloud Jenkins <svc.cloudjenkins@instructure.com>
Reviewed-by: Derek Bender <djbender@instructure.com>
Reviewed-by: Jacob Powell <spowell@instructure.com>
QA-Review: James Butters <jbutters@instructure.com>
Product-Review: James Butters <jbutters@instructure.com>
Rex Fleischer