DNM: rgw: get usage rest api now return capacity used by each bucket #8043
Conversation
|
@yehudasa prospectively do you see any issue with this? |
|
@chenji-kael , can you open a tracker issue for this? |
|
@chenji-kael , can you add it to commit message: |
chenji-kael
force-pushed
the
usage
branch
2 times, most recently
from
5c85df8
to
af3828e
Compare
|
@oritwas done |
| @@ -29,8 +29,6 @@ static const char *signed_subresources[] = { | |||
| "torrent", | |||
| "uploadId", | |||
| "uploads", | |||
| "start-date", | |||
There was a problem hiding this comment.
I think you remove those by mistake?
There was a problem hiding this comment.
@oritwas
no , start-date and end-date is only parameters， should not be treated as sub-resource ， I introduced this mistake at last commit, so I think it need wipe out.
In fact I think usage should really be put here, what`s your opinion?
|
@oritwas ok |
|
@oritwas I have update this pr, please check it |
| bucket.convert(&entry.bucket); | ||
|
|
||
| map<string, struct rgw_bucket_dir_header>::iterator hiter = headers.begin(); | ||
| for (; hiter != headers.end(); ++hiter) { |
There was a problem hiding this comment.
can we use C++11 syle auto for the for loops here
chenji-kael
force-pushed
the
usage
branch
2 times, most recently
from
498504f
to
4c5d177
Compare
|
@theanalyst @oritwas is this one ready? |
|
it looks like the comments were not fixed. |
|
also it will be good to join #9495 with this pull request |
|
@oritwas @theanalyst @chenji-kael what is the status of this PR? |
|
@chenji-kael, please rebase |
Fixes:tracker.ceph.com/issues/16191 Signed-off-by: Ji Chen <insomnia@139.com>
Fixes:tracker.ceph.com/issues/16191 Signed-off-by: Ji Chen <insomnia@139.com>
|
@oritwas rebase done |
|
@oritwas ping |
|
compression support was added to radosgw recently, and |
| ldout(cct, 0) << "ERROR: could not get bucket stats: ret=" << ret << dendl; | ||
| return ret; | ||
| } | ||
| buckets_usage_map.insert(pair<string, cls_user_bucket_entry>(bucket_ent.bucket.name, entry)); |
There was a problem hiding this comment.
buckets_usage_map.emplace(bucket_ent.bucket.name, entry); should work here
|
jenkins test this please |
|
jenkins test this please (asok) |
liewegas
changed the title
|
@oritwas what should we do with this pr? |
|
The changes needed are small. |
oritwas
changed the title
|
opened #14287 with the fixes |
It`s very useful to let users know how much capacity
they have used in each bucket, Venders also need this
function to charge fee by per bucket
Signed-off-by: Ji Chen insomnia@139.com