Thanks, The crushtool didn't help me further much unless I did something crazy as you said. So I have started by just creating a new and correct rule and just change the the pools one by one to use the new rule. This seems to work fine and as far as I can see it didn't impact any user (much). Maarten van Ingen | Systems Expert | Distributed Data Processing | SURFsara | Science Park 140 | 1098 XG Amsterdam | | T +31 (0) 20 800 1300 | maarten.vaningen@surfsara.nl | https://surfsara.nl | We are ISO 27001 certified and meet the high requirements for information security. From: "Paul Emmerich" <paul.emmerich@croit.io> To: "Maarten van Ingen" <maarten.vaningen@surfsara.nl> Cc: "ceph-users" <ceph-users@ceph.io> Sent: Tuesday, 19 November, 2019 13:36:04 Subject: Re: [ceph-users] How proceed to change a crush rule and remap pg's? I don't think that there's a feasible way to do this in a controlled manner. I would just change it and trust Ceph's remapping mechanism to work properly. You could use crushtool to calculate what the new mapping is and then do something crazy with upmaps (move them manually to the new locations one by one and then remove all upmaps and change the rule)... but that's quite annoying to do and probably doesn't really help. Paul -- Paul Emmerich Looking for help with your Ceph cluster? Contact us at [ https://croit.io/ | https://croit.io ] croit GmbH Freseniusstr. 31h 81247 München [ http://www.croit.io/ | www.croit.io ] Tel: +49 89 1896585 90 On Tue, Nov 19, 2019 at 11:11 AM Maarten van Ingen < [ mailto:maarten.vaningen@surfsara.nl | maarten.vaningen@surfsara.nl ] > wrote: Hi, I have a small but impacting error in my crush rules. For unknown reasons the rules are not using host but osd to place the data and thus we have some nodes with all three copies instead of three different nodes. We noticed this when rebooting a node and a pg became stale. My crush rule: { "rule_id": 0, "rule_name": "replicated_rule", "ruleset": 0, "type": 1, "min_size": 1, "max_size": 10, "steps": [ { "op": "take", "item": -2, "item_name": "default~hdd" }, { "op": "chooseleaf_firstn", "num": 0, "type": "osd" }, { "op": "emit" } ] }, Type should be host of course. And I want to alter this and move pg's such that all is as should. How can I best proceed in correcting this issue? I do like to throttle the remapping of the data so ceph itself won't be unavailable while the data is redistributed. We are running on Mimic (13.2.6), and this environment has been installed freshly as Mimic while using ceph-ansible. Current ceph -s output: cluster: id: <<fsid> health: HEALTH_OK services: mon: 3 daemons, quorum mon01,mon02,mon03 mgr: mon01(active), standbys: mon02, mon03 mds: cephfs-2/2/2 up {0=mon03=up:active,1=mon01=up:active}, 1 up:standby osd: 502 osds: 502 up, 502 in data: pools: 18 pools, 8192 pgs objects: 28.74 M objects, 100 TiB usage: 331 TiB used, 2.3 PiB / 2.6 PiB avail pgs: 8192 active+clean Cheers, Maarten van Ingen | Systems Expert | Distributed Data Processing | SURFsara | Science Park 140 | 1098 XG Amsterdam | | T +31 (0) 20 800 1300 | [ mailto:maarten.vaningen@surfsara.nl | maarten.vaningen@surfsara.nl ] | [ https://surfsara.nl/ | https://surfsara.nl ] | We are ISO 27001 certified and meet the high requirements for information security. _______________________________________________ ceph-users mailing list -- [ mailto:ceph-users@ceph.io | ceph-users@ceph.io ] To unsubscribe send an email to [ mailto:ceph-users-leave@ceph.io | ceph-users-leave@ceph.io ]