After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 768282 - test-suite failure: ERROR: general - too few tests run (expected 98, got 2)
test-suite failure: ERROR: general - too few tests run (expected 98, got 2)
Status: RESOLVED FIXED
Product: libgdata
Classification: Platform
Component: General
0.17.x
Other Linux
: Normal normal
: ---
Assigned To: libgdata-maint
libgdata-maint
Depends on:
Blocks:
 
 
Reported: 2016-07-01 14:03 UTC by Michael Biebl
Modified: 2016-09-21 11:56 UTC
See Also:
GNOME target: ---
GNOME version: ---



Description Michael Biebl 2016-07-01 14:03:05 UTC
Version: 0.17.5
Libsoup: 2.54.1

Hi,

when running the test-suite on a Debian unstable system on amd64, I get

=================================================
   libgdata 0.17.5: gdata/tests/test-suite.log
=================================================

# TOTAL: 300
# PASS:  298
# SKIP:  0
# XFAIL: 0
# FAIL:  0
# XPASS: 0
# ERROR: 2

.. contents:: :depth: 2

ERROR: general
==============

# random seed: R02S08bf586ecfcd444562f2b23af2fa9a9d
1..98
ok 1 /comparable
PASS: general 1 /comparable
# Start of tests tests
ok 2 /tests/xml_comparison
PASS: general 2 /tests/xml_comparison
# End of tests tests
**
libgdata:ERROR:general.c:1430:test_service_network_error: assertion failed (error == (gdata-service-error-quark, 9)): Invalid request URI or header, or unsupported nonstandard parameter: Message Corrupt (gdata-service-error-quark, 2)
# Start of service tests
# > GET / HTTP/1.1
# > Soup-Debug-Timestamp: 1467381623
# > Soup-Debug: SoupSession 1 (0x1d5c140), SoupMessage 1 (0x1d58190), SoupSocket 1 (0x1d4a3f0)
# > Host: thisshouldnotexist.localhost:443
# > GData-Version: 2
# > Accept-Encoding: gzip, deflate
# > User-Agent: libgdata/0.17.5 - gzip
# > Connection: Keep-Alive
#   
# < HTTP/1.1 8 Message Corrupt
# < Soup-Debug-Timestamp: 1467381623
# < Soup-Debug: SoupMessage 1 (0x1d58190)
#   
Aborted
# libgdata:ERROR:general.c:1430:test_service_network_error: assertion failed (error == (gdata-service-error-quark, 9)): Invalid request URI or header, or unsupported nonstandard parameter: Message Corrupt (gdata-service-error-quark, 2)
ERROR: general - too few tests run (expected 98, got 2)
ERROR: general - exited with status 134 (terminated by signal 6?)
Comment 1 Philip Withnall 2016-07-12 13:18:02 UTC
I can’t reproduce this using libsoup master (which is basically identical to 2.54.1) on Fedora 23 on an amd64 system. Is this consistently reproducible for you?
Comment 2 Michael Biebl 2016-07-12 16:11:35 UTC
Yes, I can easily and reliably reproduce the problem. Please let me know if you want me to run further diagnostics, apply and test patches, etc.
Comment 3 Philip Withnall 2016-08-02 23:46:43 UTC
(In reply to Michael Biebl from comment #2)
> Yes, I can easily and reliably reproduce the problem. Please let me know if
> you want me to run further diagnostics, apply and test patches, etc.

Sorry for the slow reply; I still haven’t been able to reproduce this, which is a bit puzzling.

Would you be able to break on soup_message_set_status() and then use the `generate-core-file` command in gdb to generate a core dump to attach here so I can poke at the variables please?
Comment 4 Philip Withnall 2016-08-02 23:52:30 UTC
I guess it could be something to do with the domain resolution — do you have an odd network setup where thisshouldnotexist.localhost might actually resolve somewhere, or something like that?
Comment 5 Philip Withnall 2016-08-02 23:58:49 UTC
I just pushed this commit, which probably won’t fix the problem, but certainly can’t hurt. Tests still pass for me.

commit 7d9cec6bab94a7de67fcaf72987df4349296191a
Author: Philip Withnall <philip@tecnocode.co.uk>
Date:   Wed Aug 3 00:56:10 2016 +0100

    tests: Use correct invalid TLD rather than localhost
    
    RFC 6761 specifies an entire top-level domain which is guaranteed to
    never resolve with a DNS query: invalid. Use that instead of localhost,
    to eliminate the chance that the domain name we thought should never
    resolve, actually resolves.
    
    https://tools.ietf.org/html/rfc6761#section-6.4
    
    This might fix bug #768282.
    
    https://bugzilla.gnome.org/show_bug.cgi?id=768282

 gdata/tests/general.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
Comment 6 Philip Withnall 2016-09-20 16:15:56 UTC
Michael, does this fix the problem for you?
Comment 7 Michael Biebl 2016-09-21 11:56:11 UTC
(In reply to Philip Withnall from comment #6)
> Michael, does this fix the problem for you?

It does indeed. Thanks!