Fix Discovery provisioning tests by archanaserver · Pull Request #20638 · SatelliteQE/robottelo

archanaserver · 2026-01-15T09:12:40Z

Problem Statement

Discovery provisioning tests are failing because dhcpd is not restarted after building PXE templates in the module_discovery_sat fixture. Without the restart, dhcpd serves old cached configuration.

Solution

Added dhcpd service restart in the module_discovery_sat fixture after all discovery configuration is complete. This ensures dhcpd picks up the updated PXE boot configuration and serves the correct discovery boot files to VMs.

Related Issues

Fixes: SAT-39920

sourcery-ai

Hey - I've left some high level feedback:

Using a bare assert on sat.execute('systemctl restart dhcpd').status == 0 will raise an uninformative AssertionError; consider checking the status and raising/logging a more descriptive error (including stdout/stderr) to ease debugging when the restart fails.
If there is (or might be) a shared helper for managing systemd services in tests, consider calling that instead of invoking systemctl directly here to keep service management behavior consistent across fixtures.

Prompt for AI Agents

Please address the comments from this code review:

## Overall Comments
- Using a bare `assert` on `sat.execute('systemctl restart dhcpd').status == 0` will raise an uninformative AssertionError; consider checking the status and raising/logging a more descriptive error (including stdout/stderr) to ease debugging when the restart fails.
- If there is (or might be) a shared helper for managing systemd services in tests, consider calling that instead of invoking `systemctl` directly here to keep service management behavior consistent across fixtures.

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

_{Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.}

evgeni · 2026-01-15T12:37:58Z

+    # Restart dhcpd to pick up new PXE configuration for discovery
+    assert sat.execute('systemctl restart dhcpd').status == 0


Why would one need to restart dhcpd? the proxy should only write to existing paths and those should be dynamically loaded by dhcpd

evgeni · 2026-01-15T12:40:35Z

trigger: test-robottelo
pytest: tests/foreman/ui/test_discoveredhost.py -vv

Satellite-QE · 2026-01-15T13:09:32Z

PRT Result

Build Number: 14037
Build Status: SUCCESS
PRT Comment: pytest tests/foreman/ui/test_discoveredhost.py -vv --external-logging
Test Result : ========== 3 passed, 5 deselected, 47 warnings in 1151.82s (0:19:11) ===========

Gauravtalreja1 · 2026-01-20T06:21:50Z

trigger: test-robottelo
pytest: tests/foreman/ --component DiscoveryImage
provisioning: true

Satellite-QE · 2026-01-20T10:43:32Z

PRT Result

Build Number: 14065
Build Status: UNSTABLE
PRT Comment: pytest tests/foreman/ --component DiscoveryImage --external-logging --include-onprem-provisioning
Test Result : == 14 failed, 13 passed, 5821 deselected, 590 warnings in 15189.94s (4:13:09) ==

archanaserver · 2026-01-21T07:46:11Z

So, the issue is stale DHCP lease state, not PXE template changes, and this seems the same issue as SAT-28381, stale dynamic leases interfering with provisioning

robottelo/robottelo/host_helpers/satellite_mixins.py

Lines 386 to 389 in 036a711

    
           # Workaround SAT-28381 
        
           if self.network_type == NetworkType.IPV4: 
        
               assert self.execute('cat /dev/null > /var/lib/dhcpd/dhcpd.leases').status == 0 
        
               assert self.execute('systemctl restart dhcpd').status == 0

archanaserver · 2026-01-21T08:06:47Z

trigger: test-robottelo
pytest: tests/foreman/ --component DiscoveryImage
provisioning: true

evgeni · 2026-01-21T08:43:25Z

So, the issue is stale DHCP lease state, not PXE template changes, and this seems the same issue as SAT-28381, stale dynamic leases interfering with provisioning

Okay, that's a much better explanation. Thanks for digging it up!

I still don't like the solution much, but at least there is a "why" now :)

archanaserver · 2026-01-21T09:06:21Z

I still don't like the solution much, but at least there is a "why" now :)

well, i'm open to revisiting this, if you have any alternative approach

evgeni · 2026-01-21T09:48:44Z

Not really, no :(

Satellite-QE · 2026-01-21T12:31:35Z

PRT Result

Build Number: 14079
Build Status: UNSTABLE
PRT Comment: pytest tests/foreman/ --component DiscoveryImage --external-logging --include-onprem-provisioning
Test Result : == 11 failed, 16 passed, 5821 deselected, 602 warnings in 15359.10s (4:15:59) ==

evgeni · 2026-01-26T08:43:24Z

Should PRT have passed with this change?

archanaserver · 2026-01-27T10:04:31Z

Should PRT have passed with this change?

it should, but lemme dig more

archanaserver · 2026-02-01T20:21:22Z

trigger: test-robottelo
pytest: tests/foreman/ --component DiscoveryImage
provisioning: true

Satellite-QE · 2026-02-02T02:25:50Z

PRT Result

Build Number: 14144
Build Status: UNSTABLE
PRT Comment: pytest tests/foreman/ --component DiscoveryImage --external-logging --include-onprem-provisioning
Test Result : == 18 failed, 9 passed, 5825 deselected, 610 warnings in 21298.62s (5:54:58) ===

archanaserver · 2026-03-11T06:26:58Z

trigger: test-robottelo
pytest: tests/foreman/ --component DiscoveryImage
provisioning: true

Satellite-QE · 2026-03-11T09:48:10Z

PRT Result

Build Number: 14693
Build Status: UNSTABLE
PRT Comment: pytest tests/foreman/ --component DiscoveryImage --external-logging --include-onprem-provisioning
Test Result : = 6 failed, 15 passed, 5932 deselected, 538 warnings, 6 errors in 11540.05s (3:12:20) =

amolpati30 · 2026-03-12T13:56:45Z

Below two tests are failing for a specific reason, and a Jira ticket has already been attached to below link. I would suggest using these changes with the given tests so that at least three tests will pass, as one of the tests uses parameterization. Right now, 12 tests are failing, and 1 failure is expected. This fix will resolve 3 additional issues, so after the re-run, there should be 8 failures remaining.

tests/foreman/ui/test_discoveredhost.py/test_positive_auto_provision_host_with_rule
tests/foreman/ui/test_discoveredhost.py/test_positive_custom_provision_pxe_host

archanaserver · 2026-03-16T06:33:06Z

trigger: test-robottelo
pytest: tests/foreman/ --component DiscoveryImage
provisioning: true

devendra104 · 2026-03-16T07:22:33Z

trigger: test-robottelo
pytest: tests/foreman/ --component DiscoveryImage
provisioning: true

archanaserver · 2026-03-17T11:01:16Z

trigger: test-robottelo
pytest: tests/foreman/ --component DiscoveryImage
provisioning: true

Satellite-QE · 2026-03-17T16:39:14Z

PRT Result

Build Number: 14751
Build Status: UNSTABLE
PRT Comment: pytest tests/foreman/ --component DiscoveryImage --external-logging --include-onprem-provisioning
Test Result : = 7 failed, 19 passed, 5933 deselected, 613 warnings, 1 error in 19853.44s (5:30:53) =

nacoool · 2026-03-18T06:52:18Z

UI tests are failing mainly failing for Navigation and Non existent element.

@archanaserver you should check corresponding code at airgun repo for

navmazing._errors.NavigationTriesExceeded: Navigation failed to reach [Details] in the specified tries
and
selenium.common.exceptions.NoSuchElementException: Message: Could not find an element Locator(by='xpath', locator='//a[@id="disable-pass-btn"]'); For documentation on this error, please visit: https://www.selenium.dev/documentation/webdriver/troubleshooting/errors#no-such-element-exception

It should solve failing tests.

amolpati30 · 2026-03-18T08:22:46Z

UI tests are failing mainly failing for Navigation and Non existent element.

@archanaserver you should check corresponding code at airgun repo for

navmazing._errors.NavigationTriesExceeded: Navigation failed to reach [Details] in the specified tries and selenium.common.exceptions.NoSuchElementException: Message: Could not find an element Locator(by='xpath', locator='//a[@id="disable-pass-btn"]'); For documentation on this error, please visit: https://www.selenium.dev/documentation/webdriver/troubleshooting/errors#no-such-element-exception

It should solve failing tests.

I have already discussed the above issue on Slack, and I am currently looking into it.

archanaserver · 2026-03-19T05:25:19Z

UI tests are failing mainly failing for Navigation and Non existent element.

@archanaserver you should check corresponding code at airgun repo for

navmazing._errors.NavigationTriesExceeded: Navigation failed to reach [Details] in the specified tries and selenium.common.exceptions.NoSuchElementException: Message: Could not find an element Locator(by='xpath', locator='//a[@id="disable-pass-btn"]'); For documentation on this error, please visit: https://www.selenium.dev/documentation/webdriver/troubleshooting/errors#no-such-element-exception

It should solve failing tests.

@nacoool this seems like a different failure, can you clarify are these airgun failures pre-existing, or were they exposed by my changes? If yes then I believe fixing this airgun issue is not part of this PR scope? and we can handle this separately if this same behavior not applying to the 2 of the API failures we can see right now?

@evgeni @amolpati30 thoughts on this?

amolpati30 · 2026-03-19T07:05:40Z

UI tests are failing mainly failing for Navigation and Non existent element.
@archanaserver you should check corresponding code at airgun repo for
navmazing._errors.NavigationTriesExceeded: Navigation failed to reach [Details] in the specified tries and selenium.common.exceptions.NoSuchElementException: Message: Could not find an element Locator(by='xpath', locator='//a[@id="disable-pass-btn"]'); For documentation on this error, please visit: https://www.selenium.dev/documentation/webdriver/troubleshooting/errors#no-such-element-exception
It should solve failing tests.

@nacoool this seems like a different failure, can you clarify are these airgun failures pre-existing, or were they exposed by my changes? If yes then I believe fixing this airgun issue is not part of this PR scope? and we can handle this separately if this same behavior not applying to the 2 of the API failures we can see right now?

@evgeni @amolpati30 thoughts on this?

The failure you are referring to was encountered recently and is not caused by your changes. We can address it separately. The remaining failures can be handled later, and I do not believe they are related to your changes. Most of the tests are passing now, so I am giving my ACK.

archanaserver · 2026-03-19T08:03:12Z

This test test_positive_reboot_all_pxe_hosts is just testing reboot functionality. and It doesn't need to set hostgroup/location/organization/build. Those lines are causing the host to auto-provision before the test can reboot it, so dropping it.

archanaserver · 2026-03-19T08:12:25Z

trigger: test-robottelo
pytest: tests/foreman/ --component DiscoveryImage
provisioning: true

Satellite-QE · 2026-03-19T10:54:11Z

PRT Result

Build Number: 14773
Build Status: UNSTABLE
PRT Comment: pytest tests/foreman/ --component DiscoveryImage --external-logging --include-onprem-provisioning
Test Result : = 4 failed, 14 passed, 5947 deselected, 311 warnings, 9 errors in 8998.45s (2:29:58) =

shubhamsg199 · 2026-03-19T12:10:49Z

I see failures related to Manifester and RHSM which are unrelated to this change so merging as is

* Fix Discovery provisioning tests * Apply SAT-33477 workaround to UI tests (cherry picked from commit 9ff92d0)

archanaserver requested a review from a team as a code owner January 15, 2026 09:12

github-actions Bot assigned archanaserver Jan 15, 2026

sourcery-ai Bot reviewed Jan 15, 2026

View reviewed changes

evgeni reviewed Jan 15, 2026

View reviewed changes

Satellite-QE added the PRT-Passed Indicates that latest PRT run is passed for the PR label Jan 15, 2026

Satellite-QE added PRT-Failed Indicates that latest PRT run is failed for the PR and removed PRT-Passed Indicates that latest PRT run is passed for the PR labels Jan 20, 2026

evgeni mentioned this pull request Jan 20, 2026

Modifications in CLI and UI Libvirt CR to resolve key error #20176

Merged

archanaserver force-pushed the fix-discovery-provisioning-tests branch 2 times, most recently from 9604670 to df70b14 Compare January 21, 2026 07:45

archanaserver force-pushed the fix-discovery-provisioning-tests branch from df70b14 to 1d3ba53 Compare February 1, 2026 20:20

archanaserver force-pushed the fix-discovery-provisioning-tests branch from 940581c to d841f20 Compare March 16, 2026 06:25

archanaserver force-pushed the fix-discovery-provisioning-tests branch 2 times, most recently from e24f3ee to dbb127b Compare March 17, 2026 10:54

amolpati30 approved these changes Mar 19, 2026

View reviewed changes

evgeni approved these changes Mar 19, 2026

View reviewed changes

archanaserver added 2 commits March 19, 2026 13:31

Fix Discovery provisioning tests

e37d856

Apply SAT-33477 workaround to UI tests

6fe1c92

archanaserver force-pushed the fix-discovery-provisioning-tests branch from dbb127b to 6fe1c92 Compare March 19, 2026 08:01

shubhamsg199 approved these changes Mar 19, 2026

View reviewed changes

shubhamsg199 merged commit 9ff92d0 into SatelliteQE:master Mar 19, 2026
9 of 10 checks passed

github-actions Bot pushed a commit that referenced this pull request Mar 19, 2026

Fix Discovery provisioning tests (#20638)

01477e2

* Fix Discovery provisioning tests * Apply SAT-33477 workaround to UI tests (cherry picked from commit 9ff92d0)

github-actions Bot pushed a commit that referenced this pull request Mar 19, 2026

Fix Discovery provisioning tests (#20638)

dedcf73

* Fix Discovery provisioning tests * Apply SAT-33477 workaround to UI tests (cherry picked from commit 9ff92d0)

This was referenced Mar 19, 2026

[6.17.z] Fix Discovery provisioning tests #21061

Closed

[6.18.z] Fix Discovery provisioning tests #21062

Closed

Gauravtalreja1 pushed a commit that referenced this pull request Mar 30, 2026

Fix Discovery provisioning tests (#20638)

e05941d

* Fix Discovery provisioning tests * Apply SAT-33477 workaround to UI tests (cherry picked from commit 9ff92d0)

		# Restart dhcpd to pick up new PXE configuration for discovery
		assert sat.execute('systemctl restart dhcpd').status == 0

Uh oh!

Conversation

archanaserver commented Jan 15, 2026

Problem Statement

Solution

Related Issues

Uh oh!

sourcery-ai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

evgeni Jan 15, 2026

Choose a reason for hiding this comment

Uh oh!

evgeni commented Jan 15, 2026

Uh oh!

Satellite-QE commented Jan 15, 2026

Uh oh!

Gauravtalreja1 commented Jan 20, 2026

Uh oh!

Satellite-QE commented Jan 20, 2026

Uh oh!

archanaserver commented Jan 21, 2026

Uh oh!

archanaserver commented Jan 21, 2026

Uh oh!

evgeni commented Jan 21, 2026

Uh oh!

archanaserver commented Jan 21, 2026

Uh oh!

evgeni commented Jan 21, 2026

Uh oh!

Satellite-QE commented Jan 21, 2026

Uh oh!

evgeni commented Jan 26, 2026

Uh oh!

archanaserver commented Jan 27, 2026

Uh oh!

archanaserver commented Feb 1, 2026

Uh oh!

Satellite-QE commented Feb 2, 2026

Uh oh!

archanaserver commented Mar 11, 2026

Uh oh!

Satellite-QE commented Mar 11, 2026

Uh oh!

amolpati30 commented Mar 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

archanaserver commented Mar 16, 2026

Uh oh!

devendra104 commented Mar 16, 2026

Uh oh!

archanaserver commented Mar 17, 2026

Uh oh!

Satellite-QE commented Mar 17, 2026

Uh oh!

nacoool commented Mar 18, 2026

Uh oh!

amolpati30 commented Mar 18, 2026

Uh oh!

archanaserver commented Mar 19, 2026

Uh oh!

amolpati30 commented Mar 19, 2026

Uh oh!

archanaserver commented Mar 19, 2026

Uh oh!

archanaserver commented Mar 19, 2026

Uh oh!

Satellite-QE commented Mar 19, 2026

Uh oh!

shubhamsg199 commented Mar 19, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

amolpati30 commented Mar 12, 2026 •

edited

Loading