Bug 1314855

(In reply to Nishanth Thomas from comment #0)
> Ceph installer should provide an API to set the failure domain information

Would you mind clarifying this further? What does it mean to "set the failure domain" ?

Failure Domains:

Failure domain configuration require changes in the ceph.conf as well as the crush maps. The configuration some what similar to the crush configuration

These are different bucket types supported:
- type 9 region
- type 8 datacenter
- type 7 room
- type 6 pod
- type 5 pdu
- type 4 row
- type 3 rack
- type 2 chassis
- type 1 root

Each host can have a single or a combination of these hierarchies. say region(APAC)->datacenter(BLR)->room()->rack-chassis etc. By default all the hosts will be added to root bucket.

For USM UX flow, please have a look at https://docs.google.com/a/redhat.com/presentation/d/1MAgpVG2Fi2UtBYUhuyMScO8zObAYvhiHLp2HQ3_YpMI/edit?usp=sharing  slides 17-23

(In reply to Nishanth Thomas from comment #3)
> Failure Domains:
> 
> Failure domain configuration require changes in the ceph.conf as well as the
> crush maps. The configuration some what similar to the crush configuration
> 

ceph.conf changes should not be a problem

> These are different bucket types supported:
> - type 9 region
> - type 8 datacenter
> - type 7 room
> - type 6 pod
> - type 5 pdu
> - type 4 row
> - type 3 rack
> - type 2 chassis
> - type 1 root

Can you create a separate ticket for crush maps? That way we can track that work more specifically.


> 
> Each host can have a single or a combination of these hierarchies. say
> region(APAC)->datacenter(BLR)->room()->rack-chassis etc. By default all the
> hosts will be added to root bucket.
> 
> For USM UX flow, please have a look at
> https://docs.google.com/a/redhat.com/presentation/d/
> 1MAgpVG2Fi2UtBYUhuyMScO8zObAYvhiHLp2HQ3_YpMI/edit?usp=sharing  slides 17-23

Pull request opened to address ceph.conf configuration:

https://github.com/ceph/ceph-installer/pull/120

marking this as POST - the pull request was merged, but it addresses the ability to modify ceph.conf, not crush maps.

A separate ticket should get opened for the work needed on crush

This bug was accidentally moved from POST to MODIFIED via an error in automation, please see mmccune with any questions

Failure Domains (described in comment 3) are not part of USM 2, so is this feature really implemented in ceph-installer?

And if yes, is there any documentation we can follow to test this feature?

For both OSD and MON, a client is allowed to fully override a `ceph.conf`  via a key value mapping on sections.

Both sections document this:

http://docs.ceph.com/ceph-installer/docs/#post--api-osd-configure-
http://docs.ceph.com/ceph-installer/docs/#post--api-mon-configure-

    conf (object) – (optional) An object that maps ceph.conf sections
    (only global, mon, osd, rgw, mds allowed) to keys and values. 
    Anything defined in this mapping will override existing settings.

The feature is not documented as "failure domain" because it just implements the ability of changing ceph.conf, not dealing with anything related to crush maps

Console uses this API during installation, example:
{"address":"172.16.180.73","calamari":false,"cluster_name":"clustr1","cluster_network":"172.16.180.0/24","conf":{"global":{"osd crush update on start":false}},"fsid":"41e9a939-fc9b-4a6b-b0d6-1efedd0a4164","host":"mkudlej-usm1-mon2.os1.phx2.redhat.com","monitor_secret":"AQA7P8dWAAAAABAAH/tbiZQn/40Z8pr959UmEA==","monitors":[{"address":"172.16.180.5","host":"mkudlej-usm1-mon1.os1.phx2.redhat.com"}],"public_network":"172.16.180.0/24","redhat_storage":true}

It works in Console with packages:
ceph-ansible-1.0.5-32.el7scon.noarch
ceph-installer-1.0.14-1.el7scon.noarch
rhscon-ceph-0.0.40-1.el7scon.x86_64
rhscon-core-0.0.41-1.el7scon.x86_64
rhscon-core-selinux-0.0.41-1.el7scon.noarch
rhscon-ui-0.0.52-1.el7scon.noarch

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHEA-2016:1754