RadosError-prefixes are not recommended, instead just Err as prefix is
used.
Also, errors are constants, not variables.
For existing users, backwards compatible constants are available. These
will need to be removed in a future go-ceph release.
Signed-off-by: Niels de Vos <ndevos@redhat.com>
The cephfs function ceph_create_from_rados requires the cluster_t
value, which is current private to the Conn type. Alias the cluster_t
type and allow other pkgs to get it if needed.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
The code path where the pools result buffer was increased in size
then re-fetched was not being tested. Test it.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
Testfiy's assert and require libs have helpful functions for
checking values. Use them to reduce a bunch of boilerplate code in the
TestGetPoolByName function.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
A number of the error handling checks do not seem to need to check
for specific non-zero error returns so we should be able to
convert to getRadosError rather than a lot of boiler-plate.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
A number of the functions in rados do not use the simple error
conversion function where is obviously applies. Convert them to use it.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
There is no good reason why the internal error handling support should
be exposed outside the rados package. Make the function lower-cased and
thus private.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
When working on the previous rbd change it was determined that the
connection setup code would block forever if the ceph config was missing
or bad when running the tests. Adjust the connection setup in the rados
pkg such that the tests will not block forever.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
Keeping the _test.go code in the same package allows for the testing
of private functions. AFAIK it is more conventional as well.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
Release resources that are allocated while configuring the connection to
the cluster. rados_shutdown() should only be needed after a successful
call to rados_connect(), however if the connection has been configured
with non-default parameters, some of the parameters may be allocated
before connecting. rados_shutdown() will free the allocated resources,
even if there has not been a connection yet.
Note that the finalizers get executed during garbage collection, which
can be forced by calling runtime.GC() for testing.
Fixes: #109
Signed-off-by: Niels de Vos <ndevos@redhat.com>
- Use C errno constants in place of hardcoded ints
- Use type inference where possible (especially C.CString
- Don't explicitly specify zero values where value would be default anway
- Rewrap some unnecessarily wrapped short lines
- Use if/else one-liners where value is not needed afterwards
Signed-off-by: Daniel Swarbrick <daniel.swarbrick@gmail.com>
ioctx.read should contact the cluster even when its slice is empty so
that errors like ENOENT can be reported.
Signed-off-by: Noah Watkins <nwatkins@redhat.com>
Niels de Vos
John Mulligan
Yue Zhu
John Mulligan
CodeLingo Bot
zhangyue
Noah Watkins
Dario Maiocchi
Daniel Swarbrick
Noah Watkins