I've been using ceph for nearly a year and one of the things I ran into quite a while back was that it seems like ceph is placing copies of objects on different OSDs but sometimes those OSDs can be on the same host by default. Is that correct? I discovered this by taking down one host and having some pgs become inactive.
So I guess you could say I want my failure domain to be the host, not the OSD. How would I accomplish this? I understand it involves changing the crush map. I've been reading over http://docs.ceph.com/docs/master/rados/operations/crush-map/ and it still isn't clear to me what needs to change. I expect I need to change the default replicated_ruleset which I'm still running: $ ceph osd crush rule dump [ { "rule_id": 0, "rule_name": "replicated_ruleset", "ruleset": 0, "type": 1, "min_size": 1, "max_size": 10, "steps": [ { "op": "take", "item": -1, "item_name": "default" }, { "op": "chooseleaf_firstn", "num": 0, "type": "host" }, { "op": "emit" } ] } ] And that I need something like: ceph osd crush rule create-replicated <rule-name> <root> <failure-domain> <class> then: ceph osd pool set <pool-name> crush_rule <rule-name> but I'm not sure what the values of <root> <failure-domain> <class> would be in my situation. Maybe: ceph osd crush rule create-replicated different-host default <failure-domain> <class> but I don't know what failure-domain or class should just by inspecting my current crush map. Suggestions are greatly appreciated! -- Tracy Reed http://tracyreed.org Digital signature attached for your safety.
signature.asc
Description: PGP signature
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com