[Bug]: test failure on tm_a_mmrm; message: Inputs changed and no longer reflect the fitted model.

[llrs-roche]

What happened?

A test failed on one of the CI for version 0.9.1.9035 :

  ══ Failed tests ════════════════════════════════════════════════════════════════
  ── Failure ('test-shinytest2-tm_a_mmrm.R:517:9'): e2e - tm_a_mmrm: Validate output on different selection on method t_mmrm_lsmeans. ──
  self$get_html(".shiny-output-error-validation") is not NULL
  
  `actual` is a character vector ('<div id="teal-teal_modules-mmrm-module-mmrm_table-table_out_main" class="shiny-html-output shiny-bound-output shiny-output-error shiny-output-error-shiny.silent.error shiny-output-error-validation recalculating" width="100%" aria-live="polite">Inputs changed and no longer reflect the fitted model. Press `Fit Model` button again to re-fit model.</div>')
  `expected` is NULL
  No validation error is observed
  Backtrace:
      ▆
   1. └─app_driver$expect_no_validation_error() at test-shinytest2-tm_a_mmrm.R:517:9
   2.   └─testthat::expect_null(...)
  
  [ FAIL 1 | WARN 0 | SKIP 0 | PASS 1924 ]
  Error: Test failures

This is a new error from last week on the same CI (v0.9.1.9031), which might be good to compare with git bisect to see what caused this.

There are several bugs opened for tm_a_mrmm, by the message it could be related to the last 2:

• tm_a_mmrm plots does not changes when on changing Subject Identifier #1151
• Diagnostic plot in module tm_a_mmrm does not appear for Adjustment Method #1152
• Diagnostic plot in module tm_a_mmrm does not changes  #1153

sessionInfo()

* using R version 4.3.1 (2023-06-16)
* using platform: x86_64-pc-linux-gnu (64-bit)
* R was compiled by
    gcc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
    GNU Fortran (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
* running under: Ubuntu 22.04.3 LTS
* using session charset: UTF-8

Relevant log output

No response

Code of Conduct

• I agree to follow this project's Code of Conduct.

Contribution Guidelines

• I agree to follow this project's Contribution Guidelines.

Security Policy

• I agree to follow this project's Security Policy.

[m7pr]

This is what I see when I run the example locally

Should we click Fit model in the test?

[shajoezhu]

Yes, please. The model run take time, that is why when we changed it, we added this "click"

[vedhav]

Will keep an 👀 on the integration test that run tomorrow to see if that PR fixed the issue, else I will reopen this issue.

[vedhav]

The integration test today was successful

[m7pr]

Hey @donyunardi you mentioned integration tests were successful for the last 4 days. I see that Scheduled job is failiing for last couple of weeks https://github.com/insightsengineering/teal.modules.clinical/actions/workflows/scheduled.yaml
Which job we were planning to satisfy through this card?

[llrs-roche]

The problem with the scheduled jobs is that many exceed the 6h limit for a given job: gcc13, 6h 4s; gcc14 6h 4s; noSuggests 6h 2s.
I'll pick the integration tests as a reference

[donyunardi]

Which job we were planning to satisfy through this card?

For this card, we're trying to satisfy the integration test.

Let's open a different one for Scheduled.

[donyunardi]

Closing this issue since we no longer see integration test failure for several days.

I was about to open a new issue but it looks like there's already an existing one:
#1215

I will put a comment there to address failed Scheduled pipeline as well.