Re: What to expect on rejoining a host to cluster?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi Matt,

that I'm using re-weights does not mean I would recommend it. There seems something seriously broken with reweights, see this message https://lists.ceph.io/hyperkitty/list/ceph-users@xxxxxxx/message/E5BYQ27LRWFNT4M34OYKI2KM27Q3DUY6/ and the thread with it. I have to wait for a client update before considering to change that to upmaps.

I'm pretty sure the script https://github.com/TheJJ/ceph-balancer linked by Stefan can be configured/tweaked to look only at the fullest and maybe the emptiest OSDs to compute a moderately sized list of re-mappings that will eliminate the outliers only. What I would not recommend is to go all balanced and 95% OSD utilisation. You will see serious performance loss after some OSDs reached 80% and if you loose an OSD or host you will have to combat the fallout of deleted upmaps.

Best regards,
=================
Frank Schilder
AIT Risø Campus
Bygning 109, rum S14
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx




[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux