** Description changed:

  [ Impact ]
  
-  * An explanation of the effects of the bug on users and justification
-    for backporting the fix to the stable release.
+ The SRU for LP: #2116098 introduced a regression for jammy only whereby
+ the %U and other macros is no longer being replaced with the intended
+ values, and remain blank.
  
-  * In addition, it is helpful, but not required, to include an
-    explanation of how the upload fixes this bug.
+ It is not clear what caused the bug. A simple no-change-rebuild of the
+ same samba package that landed in jammy-updates does not show the bug.
+ Such rebuilds were attempted locally, in a PPA, with and without
+ proposed enabled, and even a launchpad archive snapshot of 2022-04-20
+ was used (the day of the jammy-updates build), but the bug was never
+ reproduced. The no-change-rebuild always works, and was even confirmed
+ by users in this bug when trying one of the PPAs I provided.
+ 
+ Therefore, in order to make progress in this bug, I opted to include an
+ autopkgtest for the issue, and just rebuild the package. If this build
+ again results in a package showing this bug, then a much more detailed
+ and time consuming analysis must be done.
  
  [ Test Plan ]
  
-  * detailed instructions how to reproduce the bug
+ The test plan consists of two parts:
+ a) Repeat the test plan from the original SRU LP: #2116098 for jammy.
  
-  * these should allow someone who is not familiar with the affected
-    package to reproduce the bug and verify that the updated package
-    fixes the problem.
- 
-  * if other testing is appropriate to perform before landing this
-    update, this should also be described here.
+ b) the autopkgtests must pass, in particular the new test for this issue: 
smbclient-macro-expansion
+  
  
  [ Where problems could occur ]
  
-  * Think about what the upload changes in the software. Imagine the
-    change is wrong or breaks something else: how would this show up?
+ It is troubling that the root cause of the bug was not found. Normally
+ we say that a no-change-rebuild should introduce no change in behavior,
+ but in this case we are counting on it to fix the bug.
  
-  * It is assumed that any SRU candidate patch is well-tested before
-    upload and has a low overall risk of regression, but it's important
-    to make the effort to think about what ''could'' happen in the event
-    of a regression.
- 
-  * This must never be "None" or "Low", or entirely an argument as to why
-    your upload is low risk.
- 
-  * This both shows the SRU team that the risks have been considered,
-    and provides guidance to testers in regression-testing the SRU.
+ To mitigate these unknowns, the test plan will repeat the verification
+ from the original SRU, to make sure we are not regressing that one with
+ this no-change-rebuild, and also has a specific test for the issue
+ present here in this bug.
  
  [ Other Info ]
  
-  * Anything else you think is useful to include
- 
-  * Make sure to explain any deviation from the norm, to save the SRU
-    reviewer from having to infer your reasoning, possibly incorrectly.
-    This should also help reduce review iterations, particularly when the
-    reason for the deviation is not obvious.
- 
-  * Anticipate questions from users, SRU, +1 maintenance, security teams
-    and the Technical Board and address these questions in advance
- 
+ Not at this time.
  
  [ Original Description ]
  
  After the upgrade from samba 2:4.15.13+dfsg-0ubuntu1.6 to 
2:4.15.13+dfsg-0ubuntu1.7
  the %U substitution in "config file" stopped working.
  
  Configuration (in /etc/samba/smb.conf):
  
    [global]
      security = user
      server role = standalone server
      config file = /etc/samba/users/%U.conf
  
  Example per-user config file:
  /etc/samba/users/user1.conf
  
    [global]
    [daten]
      path = /1data/daten
      read only = no
  
  Expected behavior:
  - On user login (e.g. smbclient //server/daten -U user1), Samba loads
    /etc/samba/users/user1.conf.
  - This worked fine with 0ubuntu1.6.
  - Also works fine again with Ubuntu 24.04 (Samba 4.19.5).
  
  Observed behavior with 0ubuntu1.7:
  - Samba does not load the per-user config file.
  - Log file shows: %U is empty, config file not loaded.
  - Even %m substitutions in log file (log.%m) are broken, resulting in 
"0.0.0.0".
  
  Steps to reproduce:
  1. Install Ubuntu 22.04 with samba 2:4.15.13+dfsg-0ubuntu1.7
  2. Use smb.conf as above.
  3. Create /etc/samba/users/user1.conf with a test share.
  4. Run: smbclient //server/daten -U user1
     → NT_STATUS_BAD_NETWORK_NAME
     → No sign of /etc/samba/users/user1.conf being loaded in the logs.
  
  Regression:
  - Works in 2:4.15.13+dfsg-0ubuntu1.6
  - Broken in 2:4.15.13+dfsg-0ubuntu1.7 (Jammy)
  - Works again in 24.04 / Samba 4.19.5
  
  Impact:
  - Prevents per-user Samba configurations (config file = 
/etc/samba/users/%U.conf)
  - Breaks existing setups with multiple TB of user shares.
  
  Workaround:
  - Downgrade to 0ubuntu1.6 and hold package.
  - Or upgrade to Ubuntu 24.04.
  
  Please fix regression in Ubuntu 22.04 package.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2120811

Title:
  config file = /etc/samba/users/%U.conf  variables like %U empty

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/samba/+bug/2120811/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to