You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Jani Averbach <ja...@jaa.iki.fi> on 2005/02/01 21:21:29 UTC
locking: ra-local-test and changes-test are failing
Hi!
FYI, following tests are failing on locking branch (r12897):
1) subversion/tests/libsvn_ra_local/ra-local-test:
...
PASS: ra-local-test 3: svn_ra_local__split_URL: syntax validation
Segmentation fault
2) subversion/tests/libsvn_fs_base/changes-test:
...
/home/jaa/svn/svn/locking/subversion/libsvn_fs_base/err.c:99: (apr_err=160014)
svn: Reference to non-existent node '1.0.2' in filesystem 'test-repo-changes-fetch-ordering'
FAIL: changes-test 5: verify ordered-ness of fetched compressed changes
BR, Jani
Backtrace for #1:
...
PASS: ra-local-test 3: svn_ra_local__split_URL: syntax validation
PASS: ra-local-test 4: svn_ra_local__split_URL: invalid host names
Program received signal SIGSEGV, Segmentation fault.
find_entry (ht=0x5, key=0x632030, klen=25, val=0x0) at /home/jaa/svn/apr/0.9.x/tables/apr_hash.c:237
237 for (hep = &ht->array[hash & ht->max], he = *hep;
(gdb) bt
#0 find_entry (ht=0x5, key=0x632030, klen=25, val=0x0) at /home/jaa/svn/apr/0.9.x/tables/apr_hash.c:237
#1 0x000000000058c84b in apr_hash_get (ht=0x5, key=0x632049, klen=0)
at /home/jaa/svn/apr/0.9.x/tables/apr_hash.c:303
#2 0x0000000000597764 in apr_pool_userdata_get (data=0x7fbfffec90, key=0x632049 "", pool=0x0)
at /home/jaa/svn/apr/0.9.x/memory/unix/apr_pools.c:1851
#3 0x00000000004e8cfc in get_xlate_handle_node (ret=0x7fbfffed30, topage=0x1 <Address 0x1 out of bounds>,
frompage=0x63202a "UTF-8", userdata_key=0x632030 "svn-utf-uton-xlate-handle", pool=0x7fbfffef70)
at /home/jaa/svn/svn/locking/subversion/libsvn_subr/utf.c:177
#4 0x00000000004e904f in get_uton_xlate_handle_node (ret=0x7fbfffed30, pool=0x7fbfffef70)
at /home/jaa/svn/svn/locking/subversion/libsvn_subr/utf.c:296
#5 0x00000000004e9c56 in svn_utf_cstring_from_utf8 (dest=0x7fbfffed70, src=0x7fbffff596 "/repos/path",
pool=0x7fbfffef70) at /home/jaa/svn/svn/locking/subversion/libsvn_subr/utf.c:666
#6 0x000000000040dc5d in svn_repos_find_root_path (path=0x7fbffff596 "/repos/path", pool=0x7fbfffef70)
at /home/jaa/svn/svn/locking/subversion/libsvn_repos/repos.c:1595
#7 0x0000000000402a17 in svn_ra_local__split_URL (repos=0x7fbfffee58, repos_url=0x7fbfffee50,
fs_path=0x7fbfffee48, URL=0x5fdc82 "file:///repos/path", pool=0x7fbfffef70)
at /home/jaa/svn/svn/locking/subversion/libsvn_ra_local/split_url.c:120
#8 0x00000000004006af in try_split_url (url=0x5fdc82 "file:///repos/path", pool=0x7fbfffef70)
at /home/jaa/svn/svn/locking/subversion/tests/libsvn_ra_local/ra-local-test.c:164
#9 0x00000000004008ad in split_url_host (msg=0x7fbfffef18, msg_only=0, pool=0x7fbfffef70)
at /home/jaa/svn/svn/locking/subversion/tests/libsvn_ra_local/ra-local-test.c:245
#10 0x0000000000400d34 in do_test_num (progname=0x7fbffff510 "ra-local-test", test_num=5, msg_only=0,
opts=0x7fbfffef70, pool=0x890508) at /home/jaa/svn/svn/locking/subversion/tests/svn_tests_main.c:156
#11 0x0000000000401246 in main (argc=1, argv=0x7fbffff228)
at /home/jaa/svn/svn/locking/subversion/tests/svn_tests_main.c:311
(gdb)
--
Jani Averbach
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: locking: ra-local-test and changes-test are failing
Posted by Branko Čibej <br...@xbc.nu>.
Jani Averbach wrote:
>On 2005-02-02 08:56+0100, Peter N. Lundblad wrote:
>
>
>>Thanks! This was caused by the last merge not taking a prototype change in
>>the C test suite into account. Fixed in r12900.
>>
>>
>
>Thanks to you! =)
>
>BTW, We have a bug in our testing framework because when something
>just segfault our test framework won't catch it. All what we get to
>the logs is:
>
>...
> START: ra-local-test
> END: ra-local-test
>...
>
>
Yah. Python won't catch the segfaoult, we'd have to wait() for the child
processes, but then the framework wouldn't be portable. Sigh...
-- Brane
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: locking: ra-local-test and changes-test are failing
Posted by "Peter N. Lundblad" <pe...@famlundblad.se>.
On Wed, 2 Feb 2005, Jani Averbach wrote:
> On 2005-02-02 08:56+0100, Peter N. Lundblad wrote:
> ...
> START: ra-local-test
> END: ra-local-test
> ...
>
> > > 2) subversion/tests/libsvn_fs_base/changes-test:
> > > ...
> > > /home/jaa/svn/svn/locking/subversion/libsvn_fs_base/err.c:99: (apr_err=160014)
> > > svn: Reference to non-existent node '1.0.2' in filesystem 'test-repo-changes-fetch-ordering'
> > > FAIL: changes-test 5: verify ordered-ness of fetched compressed changes
> > >
> > I think sussman is looking into this. I don' have a clue.
>
> If it is any help, last time when this passed, was <Jan 21, svn locking r12808: PASS>
> and it started to fail: <Jan 22, svn locking r12818: FAIL>
>
Yes, that's because the merge in r12817 enabled these tests on the
branch;) Actually, I pointed that out in the commit message, since I
didn't know what to do about it.
Thanks,
//Peter
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: locking: ra-local-test and changes-test are failing
Posted by Jani Averbach <ja...@jaa.iki.fi>.
On 2005-02-02 08:56+0100, Peter N. Lundblad wrote:
> Thanks! This was caused by the last merge not taking a prototype change in
> the C test suite into account. Fixed in r12900.
Thanks to you! =)
BTW, We have a bug in our testing framework because when something
just segfault our test framework won't catch it. All what we get to
the logs is:
...
START: ra-local-test
END: ra-local-test
...
> > 2) subversion/tests/libsvn_fs_base/changes-test:
> > ...
> > /home/jaa/svn/svn/locking/subversion/libsvn_fs_base/err.c:99: (apr_err=160014)
> > svn: Reference to non-existent node '1.0.2' in filesystem 'test-repo-changes-fetch-ordering'
> > FAIL: changes-test 5: verify ordered-ness of fetched compressed changes
> >
> I think sussman is looking into this. I don' have a clue.
If it is any help, last time when this passed, was <Jan 21, svn locking r12808: PASS>
and it started to fail: <Jan 22, svn locking r12818: FAIL>
BR, Jani
--
Jani Averbach
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: locking: ra-local-test and changes-test are failing
Posted by "Peter N. Lundblad" <pe...@famlundblad.se>.
On Tue, 1 Feb 2005, Jani Averbach wrote:
> FYI, following tests are failing on locking branch (r12897):
>
> 1) subversion/tests/libsvn_ra_local/ra-local-test:
> ...
> PASS: ra-local-test 3: svn_ra_local__split_URL: syntax validation
> Segmentation fault
>
>
Thanks! This was caused by the last merge not taking a prototype change in
the C test suite into account. Fixed in r12900.
> 2) subversion/tests/libsvn_fs_base/changes-test:
> ...
> /home/jaa/svn/svn/locking/subversion/libsvn_fs_base/err.c:99: (apr_err=160014)
> svn: Reference to non-existent node '1.0.2' in filesystem 'test-repo-changes-fetch-ordering'
> FAIL: changes-test 5: verify ordered-ness of fetched compressed changes
>
I think sussman is looking into this. I don' have a clue.
Regards,
//Peter
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org