]>
Commit | Line | Data |
---|---|---|
494da23a | 1 | >=14.2.2 |
11fdf7f2 | 2 | -------- |
a8e16298 | 3 | |
81eedcae TL |
4 | * Nautilus-based librbd clients can now open images on Jewel clusters. |
5 | ||
494da23a TL |
6 | * The RGW "num_rados_handles" has been removed. |
7 | If you were using a value of "num_rados_handles" greater than 1 | |
8 | multiply your current "objecter_inflight_ops" and | |
9 | "objecter_inflight_op_bytes" paramaeters by the old | |
10 | "num_rados_handles" to get the same throttle behavior. | |
11 | ||
81eedcae TL |
12 | 14.2.2 |
13 | ------ | |
14 | ||
15 | * The no{up,down,in,out} related commands has been revamped. | |
16 | There are now 2 ways to set the no{up,down,in,out} flags: | |
17 | the old 'ceph osd [un]set <flag>' command, which sets cluster-wide flags; | |
18 | and the new 'ceph osd [un]set-group <flags> <who>' command, | |
19 | which sets flags in batch at the granularity of any crush node, | |
20 | or device class. | |
21 | ||
22 | * RGW: radosgw-admin introduces two subcommands that allow the | |
23 | managing of expire-stale objects that might be left behind after a | |
24 | bucket reshard in earlier versions of RGW. One subcommand lists such | |
25 | objects and the other deletes them. Read the troubleshooting section | |
26 | of the dynamic resharding docs for details. |