pam/main-exec: Don't set timeout for D-Bus connection

[adombeck]

We've seen cases in e2e-tests where device auth took longer than 2 minutes to complete. When it finally succeeds, the login fails because the subsequent SetData(authenticationBrokerIDKey, ...) call fails:

failed to call com.ubuntu.authd.pam.SetData: dbus: connection closed by user LOGIN: exiting with error System error: dbus: connection closed by user

We don't to enforce a timeout for the login procedure here. If there should be a timeout, it should be the caller who enforces it - similar to the behavior of loginwhich is configured via LOGIN_TIMEOUT in /etc/login.defs.

Closes #1431
UDENG-9792

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 87.06%. Comparing base (15c4664) to head (56fcbd0).
⚠️ Report is 27 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1448      +/-   ##
==========================================
+ Coverage   87.05%   87.06%   +0.01%     
==========================================
  Files          93       93              
  Lines        6367     6365       -2     
  Branches      111      111              
==========================================
- Hits         5543     5542       -1     
+ Misses        768      767       -1     
  Partials       56       56              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[3v1n0]

Maybe we can override the default with an env variable for tests?

[adombeck]

Maybe we can override the default with an env variable for tests?

I don't see this is as a test only problem (I should probably make that clear in the commit message as well). I can also see how users performing device auth with MFA could easily take longer than 2 minutes to complete the login.

I was wondering why we have this timeout in the first place. It was added by commit aa48751 which says "so that in case of problems we don't hang indefinitely". Should it maybe be the caller's responsibility to cancel the operation when it thinks it's taking too long - similar to the behavior of login which is configured via LOGIN_TIMEOUT in /etc/login.defs?

[3v1n0]

Oh wait, this was meant to be the timeout to init the connection.

If it's not, then we should definely not have a timeout at all that controls how long the user can stay logging (or maybe just follow the system settings as said).

The point here was to wait few seconds before we had a connection with the bus only.

[adombeck]

If it's not, then we should definely not have a timeout at all

I removed it