[test-failed]: Chrome UI Functional Tests1.test/functional/apps/timelion/_expression_typeahead·js - timelion app expression typeahead "before all" hook for "should display function suggestions filtered by function name"

[liza-mae]

Version: 7.10.0
Class: Chrome UI Functional Tests1.test/functional/apps/timelion/_expression_typeahead·js
Stack Trace:

Error: retry.tryForTime timeout: Error: retry.try timeout: TimeoutError: Waiting for element to be located By(css selector, [data-test-subj="kibanaChrome"])
Wait timed out after 61226ms
   at /var/lib/jenkins/workspace/elastic+estf-cloud-kibana-tests/JOB/ossGrp1/TASK/saas_run_kibana_tests/node/linux-immutable/ci/cloud/common/build/kibana/node_modules/selenium-webdriver/lib/webdriver.js:842:17
   at process._tickCallback (internal/process/next_tick.js:68:7)
   at onFailure (test/common/services/retry/retry_for_success.ts:28:9)
   at retryForSuccess (test/common/services/retry/retry_for_success.ts:68:13)
   at onFailure (test/common/services/retry/retry_for_success.ts:28:9)
   at retryForSuccess (test/common/services/retry/retry_for_success.ts:68:13)

Other test failures:

Test Report: https://internal-ci.elastic.co/view/Stack%20Tests/job/elastic+estf-cloud-kibana-tests/845/testReport/

[elasticmachine]

Pinging @elastic/kibana-test-triage (failed-test)

[elasticmachine]

Pinging @elastic/kibana-app (Team:KibanaApp)

[stratoula]

As far as I can see when this test fails, many others fail with the same error. So I don't think it is related to the specific test.
@dmlemeshko any ideas here?

[dmlemeshko]

The failure is related to this part of code:

  class TimelionPage {
    public async initTests() {
      await kibanaServer.uiSettings.replace({
        defaultIndex: 'logstash-*',
      });

      log.debug('load kibana index');
      await esArchiver.load('timelion');

      await PageObjects.common.navigateToApp('timelion');
    }

And based on error my guess is that PageObjects.common.navigateToApp('timelion') did not properly load the page, could be a continuous spinner or just blank page. We saw this problem with security functional tests in Kibana CI, I'm not sure if the problem was identified and fixed. To conclude, it has nothing to do with particular test (Timelion functionality), but can be related to test order execution or Kibana low performance during execution.

[dmlemeshko]

@liza-mae could you include screenshots to test failures? it should help understanding the place and reason.

[stratoula]

Thank you @dmlemeshko, from what I can see from Jenkins we get this

And in the same build another test fails with:

[dmlemeshko]

Ok, then it means we need to update suite to add user proper permissions. @rashmivkulkarni could help here. Since it works fine on Kibana CI, changing tests order could be a reason.

[LeeDr]

Looking at the Jenkins job mentioned in the description, there's a ton of 403 errors being logged and other tests failing before it gets to this one. The issue could be related to role changes. The OSS tests by default are run with the roles defined here; https://github.com/elastic/kibana/blob/7.10/test/functional/config.js#L292

defaultRoles: ['test_logstash_reader', 'kibana_admin'],

It could be that we need to change that kibana_admin role to use Kibana feature and space privileges.

@rashmivkulkarni is starting a local test.

[rashmivkulkarni]

prior test failed to save a query- here is the screenshot:

[rashmivkulkarni]

local test run againsta a 7.10 cloud instance - running exclusive test - expression_typeahead.js with the tags --include-tag=ciGroup1 --exclude-tag=skipCloud - passes .

when the entire ciGroup1 was run on 7.10 instance of cloud.

[LeeDr]

Still not sure what the problem is on this test. The index file for _expression_typeahead loads logstash_functional test data and sets the default index pattern to logstash-*. And the test calls timelion.initTests() which only loads a Kibana index, so the default roles of
defaultRoles: ['test_logstash_reader', 'kibana_admin'], should be fine. It really seems like something broke before this test and we need to work our way down the list in the order they're failing.

[stratoula]

Thank you @LeeDr and @rashmivkulkarni for looking into this. 🙂 It is a very interesting case

[LeeDr]

@rashmi worked on another test related to permissions and this could be a similar case. The test created a user with dashboard read or dashboard all privs and yet when it opened a saved dashboard it would fail because it couldn't update a saved object. The saved object it was trying to update was the index pattern. Apparently index patterns have a flag (or lack of a flag) and they need to be updated on first use, which requires different privs.
And we think that update also fails occasionally #81457 (comment)

I also found a test which was deleting all indices including the security index! This of course breaks all tests that come after it. I have a PR to fix that test here; #82736

[liza-mae]

Fixed