ceph-osd pg_num is not updated when new cepf-osd's added to ceph cluster
Bug #1394310 reported by
Dima Shulyak
This bug affects 1 person
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| Fuel for OpenStack |
Opinion
|
Low
|
Fuel Python (Deprecated) | ||
Bug Description
Ceph osd pool is not updated with new pg_num after new nodes added to ceph cluster.
Our orchestration doesnot support reconfiguration of other nodes, with dependency between already configured nodes.
Revision history for this message
| Dmitry Borodaenko (angdraug) wrote | #1 |
| Changed in fuel: | |
| status: | New → Opinion |
| importance: | High → Low |
To post a comment you must log in.
I don't think this is a bug. Fuel only sets the default_pg_num, that's only used when creating new pools. Changing after the pools for OpenStack are created has no effect on existing pools.
Also definitely not a High priority: fine-tuning performance parameters for specific cluster configuration, layout, and expected workloads is not something we should attempt to cover with heuristics in Fuel. Giving a reasonable initial default is the best we can do.