1
0
mirror of https://github.com/samba-team/samba.git synced 2025-01-29 21:47:30 +03:00
Martin Schwenke 434f9e8594 eventscripts: 60.nfs uses nfs_check_rpc_services() to check NFS RPC services
* New directory nfs-rpc-checks.d/ replaces hardcoded rules in 60.nfs

* Installation and packaging additions to handle nfs-rpc-checks.d/

* Unit test updates, including deleting 1 test that sanity checked
  test infrastructure

* Test infrastructure changes to use nfs-rpc-checks.d/

Note that this removes support for $CTDB_NFS_SKIP_KNFSD_ALIVE_CHECK in
60.nfs.  To get the equivalent behaviour, edit 20.nfsd.check and
remove/comment all lines.

Signed-off-by: Martin Schwenke <martin@meltin.net>

(This used to be ctdb commit 7e792d6768d9ca420ce3713cb122e63afd594b15)
2013-05-07 12:55:09 +10:00
..

eventscript unit tests
======================

This directory contains some eventscript unit tests for CTDB.  These
tests can be run as a non-privileged user.  There are a lot of stub
implementations of commands (located in stubs/) used to make the
eventscripts think they're running against a real system.

Test case filenames look like:

  <eventscript>.<event>.NNN.sh

The test helper functions will run <eventscript> with specified
options.  If using the simple_test() or iterate_test() helper
functions then the 1st <event> argument is automatically passed.  When
simple_test_event() is used the event name must be explicitly passed
as the 1st argument - this is more flexible and supports multiple
events per test.

Examples:

* ../run_tests.sh .

  Run all tests, displaying minimal output.

* ../run_tests.sh -s .

  Run all tests, displaying minimal output and a summary.

* ../run_tests.sh -s ./10.interface.*.sh

  Run all the tests against the 10.interface eventscript.

* ../run_tests.sh -v -s .

  Run all tests, displaying extra output and a summary.

* ../run_tests.sh -sq .

  Run all tests, displaying only a summary.

* ../run_tests.sh -X ./10.interface.startup.002.sh

  Run a test and have the eventscript itself run with "sh -x".  This
  will usually make a test fail because the (undesirable) trace output
  will be included with the output of the eventscript.  However, this
  is useful for finding out why a test might be failing.