Details
-
Improvement
-
Status: Resolved (View Workflow)
-
Major
-
Resolution: Fixed
-
None
Description
It's a pretty old behavior, but I think it makes sense to fix it at some point:
1) Setup the non-existent folder as a workspace for JNLP slaves
2) Start slave from the user who has no permissions to create folders in the path
3) Start JNLP slave
Expected result:
1) Slave startup fails (like in SSH slaves)
Actual result:
1) Slave starts
2) Slaves accepts the builds
3) The builds fail
Attachments
Issue Links
- depends on
-
JENKINS-39370 Introduce the remoting working directory concept
-
- Resolved
-
- links to
Activity
| Field | Original Value | New Value |
|---|---|---|
| Assignee | Oleg Nenashev [ oleg_nenashev ] |
| Status | Open [ 1 ] | In Progress [ 3 ] |
| Epic Link | JENKINS-38833 [ 175240 ] |
| Link |
This issue depends on |
| Remote Link | This issue links to "remoting/pull/129 (Web Link)" [ 15090 ] |
| Status | In Progress [ 3 ] | In Review [ 10005 ] |
| Summary | Jenkins JNLP agents should not connect/schedule builds if the workspace cannot be created | Jenkins JNLP agents should not connect/schedule builds if the workspace cannot be created or not writable |
| Resolution | Fixed [ 1 ] | |
| Status | In Review [ 10005 ] | Resolved [ 5 ] |
| Remote Link | This issue links to "CloudBees Internal OSS-553 (Web Link)" [ 18877 ] |