]> git.proxmox.com Git - ceph.git/blob - ceph/doc/mgr/mds_autoscaler.rst
39dadd8829298ac25884c29374efd879c2459883
[ceph.git] / ceph / doc / mgr / mds_autoscaler.rst
1 MDS Autoscaler Module
2 =====================
3
4 The MDS Autoscaler Module monitors ``fsmap`` update notifications from the mgr
5 daemon and takes action to spawn or kill MDS daemons for a file-system as per
6 changes to the:
7
8 - ``max_mds`` config value
9 - ``standby_count_wanted`` config value
10 - standby promotions to active MDS state in case of active MDS rank death
11
12 Bumping up the ``max_mds`` config option value causes a standby mds to be promoted
13 to hold an active rank. This leads to a drop in standby mds count. The MDS
14 Autoscaler module detects this deficit and the orchestrator module is notified
15 about the required MDS count. The orchestrator back-end then takes necessary
16 measures to spawn standby MDSs.
17
18 Dropping the ``max_mds`` config option causes the orchestrator back-end to kill
19 standby mds to achieve the new reduced count. Preferably standby mds are chosen
20 to be killed when the ``max_mds`` count is dropped.
21
22 An increment and decrement of the ``standby_count_wanted`` config option value
23 has a similar effect on the total MDS count. The orchestrator is notified about
24 the change and necessary action to spawn or kill standby MDSs is taken.
25
26 A death of an active MDS rank also causes promotion of a standby mds to occupy
27 the required active rank. The MDS Autoscaler notices the change in the standby
28 mds count and a message is passed to the orchestrator to maintain the necessary
29 MDS count.
30
31 NOTE: There is no CLI associated with the MDS Autoscaler Module.