buildlog contains a mix of two different builds

Bug #705342 reported by Jani Monoses
18
This bug affects 2 people
Affects Status Importance Assigned to Milestone
Launchpad itself
Fix Released
Critical
Julian Edwards

Bug Description

ants FTBFS on armel (failed upload)
https://launchpad.net/ubuntu/+source/ants/1.9+svn532-5/+buildjob/2004367
the build and upload logs however are apparently those of the package xpdf

Related branches

Jani Monoses (jani)
affects: launchpad → launchpad-buildd
Changed in launchpad-buildd:
status: New → Triaged
importance: Undecided → Critical
affects: launchpad-buildd → launchpad
summary: - incorrect buildlog for package
+ buildlog for FTBFS (failed upload) refers to a different package
Revision history for this message
Michael Bienia (geser) wrote : Re: buildlog for FTBFS (failed upload) refers to a different package

Another build: https://launchpad.net/ubuntu/+source/clisp/1:2.48-3/+buildjob/2004687

When one looks more closely on those build logs, you will see that it's a mix of two different builds (in two diffreent chroots).

summary: - buildlog for FTBFS (failed upload) refers to a different package
+ buildlog contains a mix of two different builds
Revision history for this message
Julian Edwards (julian-edwards) wrote :

It's always on the 'adoxaceae' builder.

Revision history for this message
Michael Bienia (geser) wrote :

Three more:
https://launchpad.net/ubuntu/+archive/test-rebuild-20110114-arm/+buildjob/2170418 (libcaca)
https://launchpad.net/ubuntu/+archive/test-rebuild-20110114-arm/+buildjob/2170494 (libgphoto2)
https://launchpad.net/ubuntu/+archive/test-rebuild-20110114-arm/+buildjob/2170420 (libcanberra)

The first log chains seems to start with libcaca whichs get continued in the libcanberra log which gets continues in the clisp log.

Revision history for this message
Julian Edwards (julian-edwards) wrote :
Revision history for this message
Michael Bienia (geser) wrote :
Revision history for this message
Julian Edwards (julian-edwards) wrote :

This is most likely to be when trying to abort a builder that just timed out - abort does not work on non-virtual builders.

Revision history for this message
Julian Edwards (julian-edwards) wrote :

From the log file:

2011-01-27 14:24:35+0000 [QueryProtocol,client] Builder 'satinash' being cleaned
up from ABORTED
2011-01-27 14:24:35+0000 [QueryProtocol,client] startBuild(http://satinash.buildd
:8221/, haskell-curl, 1.3.5-4ubuntu2, Release)

Changed in launchpad:
assignee: nobody → Julian Edwards (julian-edwards)
status: Triaged → In Progress
Revision history for this message
Julian Edwards (julian-edwards) wrote :

This is only happening when a buildd-admin kills an in-progress build on a builder. However, the admin is not killing everything he should so we can put the problems down to pebkac to some extent. The manager should protect against this scenario nonetheless and not allow a 2nd build to be dispatched when the builder is in certain states.

Revision history for this message
Launchpad QA Bot (lpqabot) wrote :
Changed in launchpad:
milestone: none → 11.02
tags: added: qa-needstesting
Changed in launchpad:
status: In Progress → Fix Committed
tags: added: qa-untestable
removed: qa-needstesting
William Grant (wgrant)
Changed in launchpad:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.