OK - I have verified that FileAppenderPermissionsTest is flawed on OS X. It is reading /etc/groups to find a group for the user. If it doesn’t find a group then it tries to use the user’s name - which doesn’t work because there is no group with that name. On MacOS /etc/groups is only used in single user mode. It uses a directory in “normal” mode, so this method of finding a user’s group is just not going to work. I set the group to “staff” on OS X as that will typically work unless the user has done something custom.
Ralph > On Jul 14, 2017, at 8:32 AM, Apache <[email protected]> wrote: > > Yes. I did that to the rolling tests but I don't know whether the tests are > bad for the permission stuff or whether the functionality doesn't work on > MacOS, in which case it would need to be disabled. > > Ralph > >> On Jul 14, 2017, at 8:01 AM, Gary Gregory <[email protected]> wrote: >> >> I assume you will be adding JUnit Assume calls to skip these tests? >> >> Gary >> >>> On Jul 14, 2017 00:01, "Ralph Goers" <[email protected]> wrote: >>> >>> After ignoring the test causing the previous problem I am now encountering >>> errors with the new permissions test. I am disabling the >>> RollingAppenderDirect test since it also will have the same problem with >>> FileWatcher, but I don’t know the cause of the other failures. >>> >>> Ralph >>> >>> ERROR] Failures: >>> [ERROR] FileAppenderPermissionsTest.testFileUserGroupAPI:181 >>> expected:<rw-[----]--> but was:<rw-[r--r]--> >>> [ERROR] FileAppenderPermissionsTest.testFileUserGroupAPI:183 >>> expected:<[rgoers]> but was:<[staff]> >>> [ERROR] FileAppenderPermissionsTest.testFileUserGroupAPI:181 >>> expected:<rw-r[w--]--> but was:<rw-r[--r]--> >>> [ERROR] FileAppenderPermissionsTest.testFileUserGroupAPI:181 >>> expected:<rw[xrwxrwx]> but was:<rw[-r--r--]> >>> [ERROR] >>> RollingAppenderDirectWriteTempCompressedFilePatternTest.testAppender:94 >>> No temporary file created during compression >>> [ERROR] Errors: >>> [ERROR] JeroMqAppenderTest.testClientServer:70 » TestTimedOut test >>> timed out after 600... > > >
