summaryrefslogtreecommitdiffstats
path: root/test
Commit message (Collapse)AuthorAgeFilesLines
* add support for clone the local developers version of the repo to testAdrian Likins2008-02-121-1/+12
|
* Refactor the minion/server list and group stuff. Still not ideal, butAdrian Likins2008-02-121-7/+50
| | | | | | | a touch more OO. Also update unittests/etc. Did some s/servers/minions/ renaming while I was at it as well. Need to go back and finish that up.
* seeing if this resultis 0 is dumb, just see if it executesroot2008-02-051-1/+0
| | | | We do need to be smarter about this though
* use socket.getfqdn instead of gethostname for sanity sakeAdrian Likins2008-02-051-1/+2
|
* we don't really need to see the cp output I supposeAdrian Likins2008-02-041-2/+2
|
* at the moment, we don't need any func/.* files to build, so change the cp ↵Adrian Likins2008-02-041-1/+1
| | | | | | command to ignore them (and in the process, avoiding copying around a big pile of git stuff)
* add in a bit of a kluge to try to make the test-it.sh script a bit more ↵Adrian Likins2008-02-041-1/+16
| | | | | | | robust. Some magic to try to figure out where the script is running from and adjust paths accordiningly
* make TestFileTracker test cases a bit less dependent on the state of theAdrian Likins2008-02-041-4/+6
| | | | | | | filetracker setup make the rpm command test look for some package a bit more likely to be there than "func", for now, "filesystem"
* dont test for the "@blippy" group yet till I setup a way to put the rightAdrian Likins2008-02-041-4/+13
| | | | config file in place (or mock object, whatever...)
* clean up some paths so hopefully this is more path independentAdrian Likins2008-02-041-4/+4
|
* fix typos in attr defs so "nosetests -a \!intro" works againAdrian Likins2008-02-041-2/+2
| | | | aka, skipping the introspectivish test cases that get called on every module
* add two new methods to poke at everymoduleAdrian Likins2008-02-041-1/+21
| | | | | | call "list_methods" also, call "inventory" if it exists
* typo on unittest config optionAdrian Likins2008-02-041-1/+1
|
* we don't return xmlrpclib.Faults anymore, so most of the testAdrian Likins2008-02-041-2/+5
| | | | | cases were less than useful. Fix to call func.utils.is_error instead
* instead of trying to build the local tree directly, copyAdrian Likins2008-02-041-2/+16
| | | | it to a subdir and run from there
* some more path fixesAdrian Likins2008-01-291-0/+5
| | | | | | | still need a bit more automagic to make it do the right thing when invoke from different places for now, just invoke it directly from test/
* add an optional config file "test-it.conf"Adrian Likins2008-01-291-23/+31
| | | | | | make running the unittests optional remove "pound_on_threads" stuff cause it was stupid and slow
* make testing from a src checkout work againAdrian Likins2008-01-291-11/+33
| | | | | default to building from local checkout, and "make install" instead of using an rpm
* add Makefiles so "make clean" cleans up all the crapAdrian Likins2008-01-292-0/+14
|
* Merge branch 'master' of ssh://git.fedoraproject.org/git/hosted/funcMichael DeHaan2008-01-292-20/+42
|\
| * comment out the async tests at the momentAdrian Likins2008-01-291-20/+20
| |
| * some simple unittest for groupsAdrian Likins2008-01-291-0/+22
| |
* | All exceptions, async or otherwise, now come back as easily detectable ↵Michael DeHaan2008-01-291-0/+2
|/ | | | | | | signatures. Use utils.is_error(result) to determine if something is an error or isn't. Example scripts as well as func-inventory have been updated. See async_test.py for examples.
* Fixed async_test.py for all success/failure cases. Still need to tweak the ↵Michael DeHaan2008-01-241-2/+2
| | | | | | | code so that failure information for remote commands can be obtained (i.e. detals of the exceptions) without reading the remote logs. Async is weird because the information can't come back in faults (well, it can, but we don't want it to).
* Still working on async (pardon the debug output still), there's a long ways ↵Michael DeHaan2008-01-241-13/+34
| | | | | | | to go with the partial status reporting but it is getting better at actually doing the task, just a few kinks to work out in getting results reported correctly.
* Fixup to client noglobs invocation.Michael DeHaan2008-01-241-4/+4
|
* Merge branch 'master' of ssh://git.fedoraproject.org/git/hosted/funcMichael DeHaan2008-01-232-16/+28
|\
| * some unit tests for the async stuff that aren't quite workingAdrian Likins2008-01-231-1/+25
| |
| * Merge branch 'master' of ssh://git.fedoraproject.org/git/hosted/funcAdrian Likins2008-01-232-1/+5
| |\ | | | | | | | | | | | | | | | Conflicts: func/overlord/client.py
| * | restart the services at the end, since leaving them off was kind ofAdrian Likins2008-01-231-15/+3
| | | | | | | | | | | | | | | | | | annoying remove some comments
* | | Working on async error handling, lots more to do...Michael DeHaan2008-01-231-1/+6
| |/ |/| | | | | (If it hits no exceptions, returns are right, it's the partial error case to deal with next...)
* | Double-barrel asynchronous calls. Async can now occur on both sides ↵Michael DeHaan2008-01-231-0/+3
| | | | | | | | | | | | | | | | simultaneously and still appears as if there is only one "global" job id to the API caller, the minion job id's are managed in the background, complete with partial result response as things come in which should be very nice for ajaxy implication. job_status API does still need to be modified to list active jobs as well as to store the job name.
* | Merge branch 'master' of ssh://git.fedoraproject.org/git/hosted/funcMichael DeHaan2008-01-231-36/+54
|\| | | | | | | | | | | Conflicts: test/unittest/test_client.py
| * some unit tests for the sysctl moduleAdrian Likins2008-01-231-0/+10
| |
| * removed unneeded print statementAdrian Likins2008-01-181-1/+0
| |
| * do some refactoringAdrian Likins2008-01-181-35/+44
| | | | | | | | talk to localhost by default instead of "grimlock.devel.redhat.com"
* | Moving the async and multiprocess stuff to top level so we can use themMichael DeHaan2008-01-232-2/+3
|/ | | | on the minion for minion side async funness.
* add attributes to the automagic introspection tests so we can turn them offAdrian Likins2008-01-171-0/+7
| | | | | | easy if we want, aka nosetests -a \!intro
* add methods to BaseTest to also test the introspection methodsAdrian Likins2008-01-171-1/+72
| | | | test modules need to set a "module" attr with the module name now.
* add tests for process and service modulesAdrian Likins2008-01-171-0/+34
|
* add some test cases for the mount and networktest modulesAdrian Likins2008-01-171-0/+33
|
* more test casesAdrian Likins2008-01-171-6/+63
|
* run unittests and async_test.py as part of test-it.shAdrian Likins2008-01-161-0/+16
|
* a start at some unittest modules.Adrian Likins2008-01-161-0/+65
| | | | | This expects the "python-nose" package to be installed. To test, run "nosetests" in the unittest subdir
* add a klugey sleep to make the thread pokeryjiggery a little sanerAdrian Likins2008-01-161-0/+3
|
* Jobthing is now functional (see async_test.py for example usage), we still needMichael DeHaan2008-01-151-0/+38
| | | | to delete jobs that have expired after a certain amount of time to avoid keeping too many results around in storage.
* add some test to pound on the threaded xmlrpc server a little bitAdrian Likins2008-01-131-0/+22
|
* some rudimentary "poke at the server bits"Adrian Likins2007-12-181-0/+37
| | | | we could use some unit tests for the modules, and run them from here
* clean up /var/lib/func/certmaster as wellAdrian Likins2007-12-181-3/+9
| | | | sleep a bit after starting up funcd so the certs show up in certmaster-ca -l
* added some more testing, starting to test certmaster/certmaster-ca/funcdAdrian Likins2007-12-171-0/+50
| | | | | which doesn't actually work in the test script at the moment, not sure why
generated by cgit (git 2.34.1) at 2024-05-01 02:12:52 +0000