
By storing the functional tests configuration files in the repo, we can now run the functional_tests.sh to setup, run the functional tests, and teardown. Most likely this will be able to be run as a user from the same directory as the repo, but at the moment, the configuration files are copied to /etc/swift. The only requirements are: 1. /etc/swift does not exist. That way the tests will not interfere with an existing deployment. 2. /mnt/gluster-object/test and /mnt/gluster-object/test2 must have been created and setup correctly on an XFS or GlusterFS volume 3. sudo rights without password prompt 4. glusterfs-openstack-swift-* rpm must not be installed on the system Once the requirements are met, you can execute the tests as follows: $ bash tools/functional_tests.sh Change-Id: Icdbcd420355b02e64f294df7298a3e473b343655 Signed-off-by: Luis Pabon <lpabon@redhat.com> Reviewed-on: http://review.gluster.org/5281 Reviewed-by: Peter Portante <pportant@redhat.com>
18 lines
762 B
Plaintext
18 lines
762 B
Plaintext
[DEFAULT]
|
|
# IP address of a GlusterFS volume server member. By default, we assume the
|
|
# local host.
|
|
mount_ip = localhost
|
|
|
|
# By default it is assumed the Gluster volumes can be accessed using other
|
|
# methods besides UFO (not object only), which disables a caching
|
|
# optimizations in order to keep in sync with file system changes.
|
|
object_only = yes
|
|
|
|
# Performance optimization parameter. When turned off, the filesystem will
|
|
# see a reduced number of stat calls, resulting in substantially faster
|
|
# response time for GET and HEAD container requests on containers with large
|
|
# numbers of objects, at the expense of an accurate count of combined bytes
|
|
# used by all objects in the container. For most installations "off" works
|
|
# fine.
|
|
accurate_size_in_listing = on
|