|
Hello,
When storing large, multipart objects in the Ceph Object Store (~100 GB and more), we have noticed that HEAD calls against the rados gateway for these objects are excessively slow – in fact, they are about the same as doing a GET on the object. Looking
at the logs while this is occurring, we have noticed that Ceph seems to be doing work on each individual part during these calls, and that these operations take time. As this call is only essentially getting the metadata about an object, it seems like this
should be a quick operation. Is there some kind of caching that can be enabled for object metadata that isn’t by default? Is this expected or unusual behavior? We were doing some research and were having a difficult time finding similar problems online.
We are using CentOS 7 and Ceph Version 0.80.8.
Any insight would be appreciated! Thanks!
- Eric
|
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
