[ceph-users] How to replace a failed OSD
Hi, What is the easiest way to replace a failed disk / OSD. It looks like the documentation here is not really compatible with ceph_deploy: http://ceph.com/docs/master/rados/operations/add-or-rm-osds/ It is talking about adding stuff to the ceph.conf while ceph_deploy works in a different way. (I've tried it without adding to ceph.conf and that obviously did not work) Is there a easy way to replace a single failed OSD which has been deployed with ceph_deploy? You could remove the OSD and add a new one but I would prefer to just reuse the current config / OSD numbers. Basically I would like to do a partition/format and some ceph commands to get stuff working again... Thx, Robert van Leeuwen ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] How to replace a failed OSD
Hi Robert, What is the easiest way to replace a failed disk / OSD. It looks like the documentation here is not really compatible with ceph_deploy: http://ceph.com/docs/master/rados/operations/add-or-rm-osds/ I found the following thread useful: http://www.spinics.net/lists/ceph-users/msg05854.html best regards Danny smime.p7s Description: S/MIME cryptographic signature ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] How to replace a failed OSD
On 20/11/13 22:27, Robert van Leeuwen wrote: Hi, What is the easiest way to replace a failed disk / OSD. It looks like the documentation here is not really compatible with ceph_deploy: http://ceph.com/docs/master/rados/operations/add-or-rm-osds/ It is talking about adding stuff to the ceph.conf while ceph_deploy works in a different way. (I've tried it without adding to ceph.conf and that obviously did not work) Is there a easy way to replace a single failed OSD which has been deployed with ceph_deploy? You could remove the OSD and add a new one but I would prefer to just reuse the current config / OSD numbers. Basically I would like to do a partition/format and some ceph commands to get stuff working again... I looked at this a while back (experimenting with how to add a failed osd back in after rebuilding it). Unfortunately I can't find my notes right now - but I recall the key features being (following the docs for removing an osd): - checking it is stopped/failed - remove it from the crushmap (ceph osd crush remove) - delete its auth key (ceph auth del) - remove the osd (ceph osd rm) Then recreate it using cep-deploy (osd create of similar), would (usually? reuse the (now) vacant osd num slot. Nothing needed for editing ceph.conf I recall. I'll take a better look at this tomorrow (unless someone who knows this better - which is probably a big list - corrects my musings)! Cheers Mark ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] How to replace a failed OSD
Hello, It would be great to have a command like : ceph-deply out osd.xx Physically change the drive, then ceph-deploy replace osd.xx What do you think ? Best Regards - Cordialement Alexis 2013/11/20 Mark Kirkwood mark.kirkw...@catalyst.net.nz: On 20/11/13 22:27, Robert van Leeuwen wrote: Hi, What is the easiest way to replace a failed disk / OSD. It looks like the documentation here is not really compatible with ceph_deploy: http://ceph.com/docs/master/rados/operations/add-or-rm-osds/ It is talking about adding stuff to the ceph.conf while ceph_deploy works in a different way. (I've tried it without adding to ceph.conf and that obviously did not work) Is there a easy way to replace a single failed OSD which has been deployed with ceph_deploy? You could remove the OSD and add a new one but I would prefer to just reuse the current config / OSD numbers. Basically I would like to do a partition/format and some ceph commands to get stuff working again... I looked at this a while back (experimenting with how to add a failed osd back in after rebuilding it). Unfortunately I can't find my notes right now - but I recall the key features being (following the docs for removing an osd): - checking it is stopped/failed - remove it from the crushmap (ceph osd crush remove) - delete its auth key (ceph auth del) - remove the osd (ceph osd rm) Then recreate it using cep-deploy (osd create of similar), would (usually? reuse the (now) vacant osd num slot. Nothing needed for editing ceph.conf I recall. I'll take a better look at this tomorrow (unless someone who knows this better - which is probably a big list - corrects my musings)! Cheers Mark ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] How to replace a failed OSD
Hi, Let say disk /dev/sdb failed on node nodeA. I would hot remove it, plug a new one and ceph-deploy osd create nodeA:/dev/sdb There is more context about how this is actually managed by ceph and the operating system in http://dachary.org/?p=2428 Fully automated disks life cycle in a Ceph cluster Cheers On 20/11/2013 10:27, Robert van Leeuwen wrote: Hi, What is the easiest way to replace a failed disk / OSD. It looks like the documentation here is not really compatible with ceph_deploy: http://ceph.com/docs/master/rados/operations/add-or-rm-osds/ It is talking about adding stuff to the ceph.conf while ceph_deploy works in a different way. (I've tried it without adding to ceph.conf and that obviously did not work) Is there a easy way to replace a single failed OSD which has been deployed with ceph_deploy? You could remove the OSD and add a new one but I would prefer to just reuse the current config / OSD numbers. Basically I would like to do a partition/format and some ceph commands to get stuff working again... Thx, Robert van Leeuwen ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com -- Loïc Dachary, Artisan Logiciel Libre ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com