Feature #400
mkcephfs: work on multiple nodes without using ssh
% Done:
0%
Source:
Tags:
Backport:
Reviewed:
Affected Versions:
Pull request ID:
Description
Probably two phases:
- first generate monmap, fsid, keys, etc. tar it up or something.
- then run the node-specific parts on each node using above tar.
History
#1 Updated by Sage Weil over 13 years ago
- Estimated time set to 3.00 h
- Source set to 2
#2 Updated by Sage Weil about 13 years ago
- Target version changed from 12 to v0.26
- translation missing: en.field_position deleted (
321) - translation missing: en.field_position set to 1
#3 Updated by Sage Weil about 13 years ago
After talking with TV about this, I think this should look more like:
- run node-specific steps on every host. generate crypto key and put it some specific location. e.g.,
[ssh host1] mkcephfs -n osd.1 -k /tmp/foo.key
- suck up the keys
scp host1:/tmp/foo.key /tmp/keys/osd.1
- do common steps (generate fsid, monmap)
- do monitors
scp -rp /tmp/keys monhost1:/tmp/keys
ssh monhost1 mkcephfs -n mon.0 --keydir /tmp/keys ...
...or something. Basically, do monitors last and avoid copying keys to hosts that don't need them.
#4 Updated by Sage Weil about 13 years ago
- Status changed from New to Resolved
commit:cda125087a00f81a9363a1e7d4aa232e2e219517
#5 Updated by Sage Weil about 13 years ago
- translation missing: en.field_position deleted (
1) - translation missing: en.field_position set to 1
#6 Updated by Sage Weil about 13 years ago
- translation missing: en.field_position deleted (
1) - translation missing: en.field_position set to 1
#7 Updated by Sage Weil about 13 years ago
- translation missing: en.field_story_points changed from 2 to 3
- translation missing: en.field_position deleted (
4) - translation missing: en.field_position set to 5