summaryrefslogtreecommitdiff
path: root/testsuite
AgeCommit message (Collapse)AuthorFilesLines
2007-05-21Got rid of the temporary "set -x" that was added for debugging.Wayne Davison1-2/+0
2007-05-21Set -x option to diagnose what is going wrong on FreeBSD.Wayne Davison1-0/+2
2007-05-20- Added more users and some groups to one of the files we're copying.Wayne Davison1-0/+7
2007-04-29Added a "sleep 1" to make certain failures more reproducible.Wayne Davison1-0/+3
2007-04-28Use the --super option for the non-fake test, just in case a systemWayne Davison1-0/+1
can chown without appearing to be root (uid 0).
2007-04-28Authorize 10.0.1.2 for one of the BSD test-farm systems.Wayne Davison1-1/+1
2007-04-26Got rid of the --no-ir kluge in this test.Wayne Davison1-2/+1
2007-04-26The --iconv option has now made it to the trunk.Wayne Davison1-0/+2
2007-04-26Changed a "mkdir -p" into a makedir.Wayne Davison1-1/+1
2007-04-24Put the files down in a subdir so that we test backup's dir-Wayne Davison1-7/+7
creating functionality (which had a bug until recently).
2007-04-24Use a single makepath call with multiple args.Wayne Davison2-5/+2
2007-04-24Allow makepath to take multiple args.Wayne Davison1-6/+6
2007-04-24Specify an arg sequence that was recently failing to work.Wayne Davison1-3/+6
2007-04-24Use a slightly different rsync calling syntax.Wayne Davison1-2/+2
2007-04-24The chown script now has a --fake-super mode if the script name hasWayne Davison1-11/+34
"fake" in it. Added a symlink named chown-fake.test.
2007-04-24Made the devices.test script do a real device test when calledWayne Davison1-10/+15
normally, or a fake-device test when called with "fake" in the script name. Added a devices-fake.test symlink.
2007-04-24Added arg-parsing to tls.c.Wayne Davison2-2/+2
2007-04-24Adding the --fake-super option.Wayne Davison1-11/+44
2007-04-18Got rid of protocol-29 check.Wayne Davison1-4/+0
2007-04-18If ssh access is not enabled, use a shell function to test thatWayne Davison1-22/+26
we can copy using a spawned rsync process.
2007-04-07Allow a local-server copy to use -X and -A even if a --protocol-versionWayne Davison2-8/+0
is forced, which makes some testing idioms easier.
2007-04-07The improved --xattrs option is landing on the trunk.Wayne Davison1-0/+107
2007-04-07Tweaked an error message and a comment.Wayne Davison1-1/+1
2007-03-18Use relative directories, and note that the merge code is onlyWayne Davison1-33/+33
working because it is not yet using --inc-recursion.
2007-03-18Added a test of a multi-file copy from a nested directory using aWayne Davison1-0/+7
relative source path.
2007-03-18One more tweak.Wayne Davison1-2/+1
2007-03-18Fixed a multi-file, no-dir, relative copy, such as:Wayne Davison1-1/+3
rsync -av lib/*.c /var/tmp/lib
2007-03-14I managed to duplicate the hang, so hopefully it's fixed now.Wayne Davison2-6/+4
Let's restore the full test suite and find out.
2007-03-14Next step: allow protocol 30 to be used, but disallow the useWayne Davison1-4/+4
of incremental recursion. If this still avoids the hanging, I will need to try to track down how inc_recurse can deadlock.
2007-03-12Next step: see if the hang returns with a partial restoration ofWayne Davison1-1/+1
protocol 30.
2007-03-12It looks like some of the compile-farm hosts are hanging on this test,Wayne Davison1-5/+5
so let's try some remote-diagnostics (let's try protocol 29 first).
2007-03-11Temporarily skip this test as it is taking too long on theWayne Davison1-0/+2
compile farm at the moment.
2007-03-11The ACL support has arrived! This version has a brand new protocolWayne Davison4-80/+189
that makes it incompatible with all prior versions. A patch will be provided to allow talking with older (patched) rsync versions.
2006-12-29Added a sleep so that the --delete-delay action will always tweakWayne Davison1-0/+1
the directory time, and reveal any lack of dir-time touch-up.
2006-12-28- Added a sleep to make sure that the preservation of the mtime onWayne Davison1-1/+1
implied directories is working.
2006-12-28Handle slight differences in the output of protocol 30's incremental mode.Wayne Davison1-19/+30
2006-12-28Use --delete-delay instead of --delete-after.Wayne Davison1-1/+1
2006-12-26Test --only-write-batch option.Wayne Davison1-0/+6
2006-12-16Make the error messages a little clearer.Wayne Davison1-3/+3
2006-12-16Changed the test for the 2.6.8 bug to preserve owner and group.Wayne Davison1-1/+1
2006-12-09Handle the new output from the hard-link code, including a fixWayne Davison1-6/+4
for a long-standing bug combinding -H with --compare-dest.
2006-12-08Use --delete-delay instead of --delete-after.Wayne Davison1-1/+1
2006-11-22Switch over to testing --remove-source-files, not *-sent-*.Wayne Davison1-2/+2
2006-11-10Test --del too.Wayne Davison1-1/+8
2006-11-10Fixed the hard-linking of symlinks test (we need to use a symlinkWayne Davison1-2/+7
to a non-existent file for the test to work right).
2006-11-10Changed how we deal with the alternate-dest directory becauseWayne Davison1-21/+20
some systems change the time on a directory that gets moved.
2006-11-10If the OS can't hard-link a symlink, tweak the expected outputWayne Davison1-1/+5
when using --link-dest.
2006-11-10Changed to reflect new (more correct) output of what updating isWayne Davison1-41/+25
happening.
2006-11-10Improved to test --link-dest (when hard-linking of devices works).Wayne Davison1-1/+19
2006-10-13Some OSes can't chmod -t for a file, so I changed the logic of ourWayne Davison1-3/+3
to to substitute a chmod that will work everywhere.