Note: This is a public test instance of Red Hat Bugzilla. The data contained within is a snapshot of the live data so any changes you make will not be reflected in the production Bugzilla. Email is disabled so feel free to test any aspect of the site that you want. File any problems you find or give feedback at bugzilla.redhat.com.
Bug 1041279 - Failed WAL and File I/O tests on sparc and ppc architectures
Summary: Failed WAL and File I/O tests on sparc and ppc architectures
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: Red Hat Enterprise Linux 7
Classification: Red Hat
Component: sqlite
Version: 7.0
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: rc
: ---
Assignee: Jan Staněk
QA Contact: qe-baseos-daemons
URL:
Whiteboard:
Depends On:
Blocks: 1041302
TreeView+ depends on / blocked
 
Reported: 2013-12-12 14:14 UTC by Jan Staněk
Modified: 2014-01-16 09:18 UTC (History)
2 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
: 1041302 (view as bug list)
Environment:
Last Closed: 2014-01-16 09:18:30 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)
PPC build.log (5.87 MB, text/x-log)
2013-12-12 14:14 UTC, Jan Staněk
no flags Details

Description Jan Staněk 2013-12-12 14:14:48 UTC
Created attachment 835814 [details]
PPC build.log

Description of problem:
During the last rpmdiff check, it was discovered that some tests from the TCL test suite fails. However, the compilation does not terminate.

By going back few releases, I found that the tests was failing for some time without notice.

Version-Release number of selected component (if applicable):
sqlite-3.7.17-2.el7 and older

How reproducible:
Build sqlite on ppc, ppc64, s390 or s390x.

Actual results:
Failures on these tests: fts3conf-3.1 fts3conf-3.2 fts3conf-3.3 fts3conf-3.4 fts3conf-3.5 fts3conf-3.6 fts3conf-3.8 wal-13.2.2 wal-13.2.3 wal-13.3.13.a wal-13.3.13.b wal-13.3.13.c wal-13.3.14.a wal-13.3.14.b wal-13.3.14.c wal-13.3.15.a wal-13.3.15.b wal-13.3.15.c wal-13.4.13.a wal-13.4.13.b wal-13.4.13.c wal-13.4.14.a wal-13.4.14.b wal-13.4.14.c wal-13.4.15.a wal-13.4.15.b wal-13.4.15.c wal-20.2 wal-20.3 wal-20.4 wal2-8.1.2 wal2-8.1.3 wal2-8.1.4 wal3-1.1.1 wal3-1.1.5 wal3-1.1.7 wal3-1.2.1 wal3-1.2.5 wal3-1.2.7 wal3-1.3.1 wal3-1.3.5 wal3-1.3.7 wal3-1.4.1 wal3-1.4.5 wal3-1.4.7 wal3-1.5.1 wal3-1.5.5 wal3-1.5.7 wal3-1.6.1 wal3-1.6.5 wal3-1.6.7 wal3-1.7.1 wal3-1.7.3 wal3-1.7.5 wal3-1.7.6 wal3-1.7.7 wal3-1.8.1 wal3-1.8.3 wal3-1.8.5 wal3-1.8.6 wal3-1.8.7 wal3-1.9.1 wal3-1.9.3 wal3-1.9.5 wal3-1.9.6 wal3-1.9.7 wal3-1.10.1 wal3-1.10.3 wal3-1.10.5 wal3-1.10.6 wal3-1.10.7 wal3-1.11.1 wal3-1.11.3 wal3-1.11.5 wal3-1.11.6 wal3-1.11.7 wal3-1.12.1 wal3-1.12.3 wal3-1.12.5 wal3-1.12.6 wal3-1.12.7 wal3-1.13.1 wal3-1.13.3 wal3-1.13.5 wal3-1.13.6 wal3-1.13.7 wal3-1.14.1 wal3-1.14.3 wal3-1.14.5 wal3-1.14.6 wal3-1.14.7 wal3-1.15.1 wal3-1.15.3 wal3-1.15.5 wal3-1.15.6 wal3-1.15.7 wal3-1.16.1 wal3-1.16.3 wal3-1.16.5 wal3-1.16.6 wal3-1.16.7 wal3-1.17.1 wal3-1.17.3 wal3-1.17.5 wal3-1.17.6 wal3-1.17.7 wal3-1.18.1 wal3-1.18.3 wal3-1.18.5 wal3-1.18.6 wal3-1.18.7 wal3-1.19.1 wal3-1.19.3 wal3-1.19.5 wal3-1.19.6 wal3-1.19.7 wal3-1.20.1 wal3-1.20.3 wal3-1.20.5 wal3-1.20.6 wal3-1.20.7 wal3-1.21.1 wal3-1.21.3 wal3-1.21.5 wal3-1.21.6 wal3-1.21.7 wal3-1.22.1 wal3-1.22.3 wal3-1.22.5 wal3-1.22.6 wal3-1.22.7 wal3-1.23.1 wal3-1.23.3 wal3-1.23.5 wal3-1.23.6 wal3-1.23.7 wal3-1.24.1 wal3-1.24.3 wal3-1.24.5 wal3-1.24.6 wal3-1.24.7 wal3-1.25.1 wal3-1.25.3 wal3-1.25.5 wal3-1.25.6 wal3-1.25.7 wal3-1.26.1 wal3-1.26.3 wal3-1.26.5 wal3-1.26.6 wal3-1.26.7 wal3-1.27.1 wal3-1.27.3 wal3-1.27.5 wal3-1.27.6 wal3-1.27.7 wal3-1.28.1 wal3-1.28.3 wal3-1.28.5 wal3-1.28.6 wal3-1.28.7 wal3-1.29.1 wal3-1.29.3 wal3-1.29.5 wal3-1.29.6 wal3-1.29.7 wal3-1.30.1 wal3-1.30.3 wal3-1.30.5 wal3-1.30.6 wal3-1.30.7 wal3-1.31.1 wal3-1.31.3 wal3-1.31.5 wal3-1.31.6 wal3-1.31.7 wal3-1.32.1 wal3-1.32.3 wal3-1.32.5 wal3-1.32.6 wal3-1.32.7 wal3-1.33.1 wal3-1.33.3 wal3-1.33.5 wal3-1.33.6 wal3-1.33.7 wal3-1.34.1 wal3-1.34.3 wal3-1.34.5 wal3-1.34.6 wal3-1.34.7 wal3-1.35.1 wal3-1.35.3 wal3-1.35.5 wal3-1.35.6 wal3-1.35.7 wal3-1.36.1 wal3-1.36.3 wal3-1.36.5 wal3-1.36.6 wal3-1.36.7 wal3-1.37.1 wal3-1.37.3 wal3-1.37.5 wal3-1.37.6 wal3-1.37.7 wal3-1.38.1 wal3-1.38.3 wal3-1.38.5 wal3-1.38.6 wal3-1.38.7 wal3-1.39.1 wal3-1.39.3 wal3-1.39.5 wal3-1.39.6 wal3-1.39.7 wal3-1.40.1 wal3-1.40.3 wal3-1.40.5 wal3-1.40.6 wal3-1.40.7 wal3-1.41.1 wal3-1.41.3 wal3-1.41.5 wal3-1.41.6 wal3-1.41.7 wal3-1.42.1 wal3-1.42.3 wal3-1.42.5 wal3-1.42.6 wal3-1.42.7 wal3-1.43.1 wal3-1.43.3 wal3-1.43.5 wal3-1.43.6 wal3-1.43.7 wal3-1.44.1 wal3-1.44.3 wal3-1.44.5 wal3-1.44.6 wal3-1.44.7 wal3-1.45.1 wal3-1.45.3 wal3-1.45.5 wal3-1.45.6 wal3-1.45.7 wal3-1.46.1 wal3-1.46.3 wal3-1.46.5 wal3-1.46.6 wal3-1.46.7 wal3-1.47.1 wal3-1.47.3 wal3-1.47.5 wal3-1.47.6 wal3-1.47.7 wal3-1.48.1 wal3-1.48.3 wal3-1.48.5 wal3-1.48.6 wal3-1.48.7 wal3-1.49.1 wal3-1.49.3 wal3-1.49.5 wal3-1.49.6 wal3-1.49.7 wal3-10.1.1 wal3-10.2.1 wal9-1.2 wal9-1.6


Expected results:
No failures

Comment 2 Honza Horak 2013-12-18 14:13:13 UTC
When we have a fix for this, we should also change the test-suite to terminate build in case some test fail.

Comment 3 Jan Staněk 2014-01-13 12:52:52 UTC
I noticed this section in specfile:

> # XXX shell tests are broken due to loading system libsqlite3, work around...    
> export LD_LIBRARY_PATH=`pwd`/.libs                                               
> export MALLOC_CHECK_=3                                                           
> %ifarch s390 s390x ppc ppc64 %{sparc} %{arm}

The "ifarched" architectures are the same that this bug manifests on (aarch64 missing, obviously). In Fedora adding the aarch64 to this list seemed to solve some build problems (see [1]).

I'm inclined to solve it the same way in RHEL - it seems that this is long-term library loading issue.

Comment 4 Jan Staněk 2014-01-13 12:55:34 UTC
Forgot to include referenced ([1] above) link to Fedora cgit:

http://pkgs.fedoraproject.org/cgit/sqlite.git/commit/?id=c79a6e8a659e33e12824fd9f614e29761b8d45bd

Comment 5 Honza Horak 2014-01-15 16:46:40 UTC
Let's add aarch64 for now and we can keep another cloned bug opened to track failing log files generally.

Comment 6 Jan Staněk 2014-01-16 09:18:30 UTC
After the last build the tests are no longer failing. Possible causes:
 - As per the specfile comment, the test mixed new and old sqlite in processing, thus failing (and any latter build with same version is OK).
 - Some dependency was fixed.

Anyway, closing this as NOTABUG (in sqlite, at least) for now, if it manifest itself in the future, I should reopen it.


Note You need to log in before you can comment on or make changes to this bug.