On Thu, Jun 29, 2017 at 1:33 PM, Gregory Farnum <gfar...@redhat.com> wrote: > I'm not sure if there are built-in tunable commands available (check the > manpages? Or Jason, do you know?), but if not you can use any generic > tooling which limits how much network traffic the RBD command can run.
Long-running RBD actions are controlled by the "rbd concurrent management ops" configuration value (default 10) which is essentially how many in-flight IOs it will send to the OSDs in parallel. I am surprised that a single rollback operation is causing a 100x increase in latency. Anything special about your environment? cache tiering in use? -- Jason _______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
