Bug 9296

Summary: Failed to install groff package on smart
Product: [Build System, Metadata & Runtime] OE-Core Reporter: Francisco Pedraza <francisco.j.pedraza.gonzalez>
Component: deploymentAssignee: Francisco Pedraza <francisco.j.pedraza.gonzalez>
Status: RESOLVED INVALID QA Contact: Francisco Pedraza <francisco.j.pedraza.gonzalez>
Severity: normal    
Priority: Medium+ CC: anibal.limon, mark.hatle, ross.burton
Version: 2.1   
Target Milestone: 2.1 M4   
Hardware: NUC   
OS: x86_64   
Whiteboard:
OS type for building Yocto: --- Type of Regression: Regression (Found by QA test)
Verified: Documentation change: No (bug/feature does not impact docs)
Attachments:
Description Flags
logs
none
useradd test fails none

Description Francisco Pedraza 2016-03-17 21:34:51 UTC
STATUS: FAILED
BUILD: 2.1_M3.RC1: 8debfea81e69d038bd2d56314b272cb74f5582ed
ENVIRONMENT: genericx86-64 on NUC
STEPS TO REPRODUCE: 
Make sure the default rpm-sys channel is enabled.
Copy a package to target.
Run "smart install /path/to/package.file". For example, "smart install /home/root/groff-1.20.1-r2.i586.rpm". 

EXPECTED OUTCOME: 
The package "groff" should be installed. 

ACTUAL OUTCOME:
The package "groff" was not installed.
Tested on LAVA.

Steps to reproduce:
1. Make sure the default rpm-sys channel is enabled.
2. Copy a package to target.
3. Run "smart install /path/to/package.file". For example, "smart install /home/root/groff-1.20.1-r2.i586.rpm".
Comment 1 Aníbal Limón 2016-03-30 22:58:34 UTC
I need the logs/output for the smart utility for see what happens.
Comment 2 Francisco Pedraza 2016-04-11 16:56:17 UTC
Created attachment 3113 [details]
logs
Comment 3 Francisco Pedraza 2016-04-11 16:58:16 UTC
I looked in to logs:
19:23:49 - bitbake-worker - RESULTS - Testcase 723:
Traceback (most recent call last):
  File /work/weekly/runtime/genericx86-64_nuc/poky/meta/lib/oeqa/utils/decorators.py, line 106, in wrapped_f
    return func(*args, **kwargs)
  File /work/weekly/runtime/genericx86-64_nuc/poky/meta/lib/oeqa/utils/decorators.py, line 94, in wrapped_f
    return f(*args, **kwargs)
  File /work/weekly/runtime/genericx86-64_nuc/poky/meta/lib/oeqa/runtime/smart.py, line 117, in test_smart_install_from_disk
    self.smart('install -y ./psplash-default*')
  File /work/weekly/runtime/genericx86-64_nuc/poky/meta/lib/oeqa/utils/decorators.py, line 94, in wrapped_f
    return f(*args, **kwargs)
  File /work/weekly/runtime/genericx86-64_nuc/poky/meta/lib/oeqa/runtime/smart.py, line 22, in smart
    self.assertEqual(status, expected, message)
AssertionError: 1 != 0 : smart install -y ./psplash-default*
Loading cache...
Updating cache...               ######################################## [100%]

Computing transaction...
error: Can't install psplash-default-0.1+git0+14c8f7b705-r15@core2_64: it can't coexist with psplash-default-0.1+git0+88343ad23c-r15@core2_64

In logs is not described, to install "groff" however in Test Case 723 is described groff package https://bugzilla.yoctoproject.org/tr_show_case.cgi?case_id=723
I am not sure if can be a miss-match between this.
Comment 4 Aníbal Limón 2016-04-11 21:22:26 UTC
Reviewing the log seems that the tests runs twice in the same hw platform that's the reason for fail see failure to add again the test1 user,


-----------------------
AssertionError: False is not true : Failed to create new user: useradd: user 'test1' already exists
-----------------------

Also there are a previous RPM installed in the platform see,

-----------------------
Computing transaction...
error: Can't install psplash-default-0.1+git0+14c8f7b705-r15@core2_64: it can't coexist with psplash-default-0.1+git0+88343ad23c-r15@core2_64
-----------------------

Paco could you run again this tests in the same revision?
Comment 5 Aníbal Limón 2016-04-12 14:44:54 UTC
Created attachment 3119 [details]
useradd test fails

Executing twice the tests over the same image fails with useradd tests.

For me the issue was the execution of the tests twice in the same hw platform.
Comment 6 Francisco Pedraza 2016-04-14 19:46:20 UTC
According with the evidence is an invalid one, we have run several times getting the same results.
The problem appears because the test was run twice over the same platforms with different image versions.