On Mon, Jan 2, 2017 at 4:25 AM, Jens Dueholm Christensen <j...@ramboll.com> wrote:
> On Friday, December 30, 2016 07:05 PM Brian Andrus wrote: > > > We have a set it and forget it cronjob setup once an hour to keep things > a bit more balanced. > > > > 1 * * * * /bin/bash /home/briana/reweight_osd.sh 2>&1 | /usr/bin/logger > -t ceph_reweight > > > > The script checks and makes sure cluster health is OK and no other > rebalancing is going on. It will > > also check the reported STDDEV from `ceph osd df` and if outside > acceptable ranges executes a > > gentle reweight. > > Would you mind sharing that script? The script attached is a quick and dirty bash script, not perfect, blah blah blah... consider it still under development but feel free to use it as a base for your own environment needs. https://gist.github.com/oddomatik/1e94c67f521ebd16c789e4cbe1d0a5d1 > The three parameters after the reweight-by-utilization are not well > documented, but they are > > > > 103 - Select OSDs that are 3% above the average (default is 120 but we > want a larger pool of OSDs to > choose from to get an eventual tighter tolerance) > > .010 - don't reweight any OSD more than this increment (keeps the impact > low) > > 10 - number of OSDs to select (to keep impact manageable) > > Ah! Thank you for that pointer. > > For the record the same arguments can be used for dry-runs of "ceph osd > test-reweight-by-utilization ..." > and correspond to these values in the output from > test-reweight-by-utilization: > > oload 120 > max_change 0.05 > max_change_osds 4 > > The above values are the current defaults in Hammer (0.94.9), but can > easily be changed to see the impact > before running the actual rebalance.. > > Regards, > Jens Dueholm Christensen > _______________________________________________ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > -- Brian Andrus Cloud Systems Engineer DreamHost, LLC
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
