In openSUSE there is a policy to use %fdupes in the spec file if RPMLINT
complains about duplicate files wasting space in the filesystem.
However, RPMLINT is not so complaining, so drop fdupes.
http://tracker.ceph.com/issues/12301Fixes: #12301
Signed-off-by: Nathan Cutler <ncutler@suse.com>
For scraping cephfs data pools for metadata in
backtraces, and injecting that metadata into
the cephfs metadata pool (initially offline).
Signed-off-by: John Spray <john.spray@redhat.com>
These are deliberately fairly sparse, because:
* These tools are for experts
* These tools may well be wrapped in a higher
level recovery tool that orchestrates parallel
workers at some stage.
Signed-off-by: John Spray <john.spray@redhat.com>
For use in cephfs-data-scan for file size and mtime
recovery, by doing a set-if-greater for size/mtime
xattrs on the 0th object of a file.
Signed-off-by: John Spray <john.spray@redhat.com>
...and correspondingly bump the cluster protocol
version as these structures are not versioned
themselves.
Signed-off-by: John Spray <john.spray@redhat.com>
Used to mark inodes+fnodes with flags indicating
that particular types of damage have either
been detected (scrub/fetch) or caused (cephfs-data-scan
inject).
This should serve as a cue to the MDS to do online
repair of stats for the nodes that have been injected.
Signed-off-by: John Spray <john.spray@redhat.com>
Because it only depends on the data in the store, not
anything from a live MDS, and it's useful when
we're dealing with fragmented directories from
an offline tool.
Signed-off-by: John Spray <john.spray@redhat.com>
When we failed and prompted for --force, it still
looked like a success. ::reset never had any
proper error handling since it was lifted out
of the MDS, so add it now.
Signed-off-by: John Spray <john.spray@redhat.com>
HTTP/1.1 assumes Keep-Alive by default, but if a Connection: close header is send
the server should respond with it as well.
This makes the client close the connection after the request.
Fixes: #12298
Currently vstart.sh only print out S3 user info. This patch updates it
to print swift user info too. Also changes the swift account/user/password
to test/tester/testing which align with openstack-swift better:
https://github.com/openstack/swift/blob/master/etc/proxy-server.conf-sample#L270
Signed-off-by: Yuan Zhou <yuan.zhou@intel.com>
remove rgw_admin.cc OPT_SUBUSER_MODIFY, show_user_info code block.
switch (opt_cmd) {
...
case OPT_SUBUSER_MODIFY:
show_user_info(info, formatter); //show first time (remove this)
break;
...
}
// output the result of a user operation
if (output_user_info) {
...
show_user_info(info, formatter); //show second time
}
test fix:
before: after subuser modify print twice user info.
after changes, do the same procedure, print only once user info.
Signed-off-by: guce guce@h3c.com
The 5GB limit of a single operation uploading was part of S3 spec.
However some private setups may have some special requirements
on this limit. It's more convinent to have a configurable value.
Fixes: #6999
Signed-off-by: Yuan Zhou <yuan.zhou@intel.com>
Improve various error messages generated with invalid syntax
Add test cases for most of these error messages
Signed-off-by: David Zafman <dzafman@redhat.com>
A valid pgid had to be specified in order for superblock to be displayed,
but it was otherwise ignored.
Fix message that didn't include dump-super
Signed-off-by: David Zafman <dzafman@redhat.com>
Conflicts:
src/osd/ReplicatedPG.cc
src/test/osd/RadosModel.h
Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: David Zafman <dzafman@redhat.com>
Add an explanatory comment to parent_types.h that explains what
the "overlap" member in parent_info means. This was not exactly easy
to figure out.
Signed-off-by: Krzysztof Kosiński <krzysztof.kosinski@intel.com>
