Re: [ceph-users] installation docs

2016-12-30 Thread Goncalo Borges
Hi Manuel I am Goncalo Borges (Portuguese) and I work at the university of Sydney. We have been using ceph and cephfs since almost two years. If you think worthwhile, we can just talk and discuss our experiences. There is good ceph community in Melbourne but you are actually the first one in Syd

[ceph-users] linux kernel version for clients

2016-12-30 Thread Manuel Sopena Ballesteros
Hi, I have several questions regarding kernel running on client machines: * Why is kernel 3.10 considered an old kernel to run ceph clients? * Which features are missing? * What would be the impact of running clients on centos7.3 (kernel 3.10) compared to using a highe

Re: [ceph-users] How to know if an object is stored in clients?

2016-12-30 Thread Nick Fisk
Just to add, the rados command will block until the objects are stored on sufficient number of OSD’s and you should also be able to check the return code to confirm that there wasn’t any errors as well. From: ceph-users [mailto:ceph-users-boun...@lists.ceph.com] On Behalf Of Jaemyoun Lee Sen

Re: [ceph-users] installation docs

2016-12-30 Thread Brad Hubbard
+ceph-devel On Fri, Dec 30, 2016 at 6:02 PM, Manuel Sopena Ballesteros wrote: > Hi, > > > > I just would like to point a couple of issues I have following the > INSTALLATION (QUICK) document. > > > > 1. The order to clean ceph deployment is: > > a. Ceph-deploy purge {ceph-node} [{ceph

Re: [ceph-users] Unbalanced OSD's

2016-12-30 Thread Shinobu Kinjo
The best practice to reweight OSDs is to run test-reweight-by-utilization which is dry-run of reweighting OSDs before running reweight-by-utilization. On Sat, Dec 31, 2016 at 3:05 AM, Brian Andrus wrote: > We have a set it and forget it cronjob setup once an hour to keep things a > bit more balan

Re: [ceph-users] Unbalanced OSD's

2016-12-30 Thread Brian Andrus
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

Re: [ceph-users] Unbalanced OSD's

2016-12-30 Thread Shinobu Kinjo
On Fri, Dec 30, 2016 at 7:27 PM, Kees Meijs wrote: > Thanks, I'll try a manual reweight at first. Great. CRUSH would probably be able to be more clever in the future anyway. > > Have a happy new year's eve (yes, I know it's a day early)! > > Regards, > Kees > > On 30-12-16 11:17, Wido den Holla

Re: [ceph-users] Unbalanced OSD's

2016-12-30 Thread Kees Meijs
Thanks, I'll try a manual reweight at first. Have a happy new year's eve (yes, I know it's a day early)! Regards, Kees On 30-12-16 11:17, Wido den Hollander wrote: > For this reason you can do a OSD reweight by running the 'ceph osd > reweight-by-utilization' command or do it manually with 'cep

Re: [ceph-users] Unbalanced OSD's

2016-12-30 Thread Shinobu Kinjo
On Fri, Dec 30, 2016 at 7:17 PM, Wido den Hollander wrote: > >> Op 30 december 2016 om 11:06 schreef Kees Meijs : >> >> >> Hi Asley, >> >> We experience (using Hammer) a similar issue. Not that I have a perfect >> solution to share, but I felt like mentioning a "me too". ;-) >> >> On a side note:

Re: [ceph-users] Unbalanced OSD's

2016-12-30 Thread Wido den Hollander
> Op 30 december 2016 om 11:06 schreef Kees Meijs : > > > Hi Asley, > > We experience (using Hammer) a similar issue. Not that I have a perfect > solution to share, but I felt like mentioning a "me too". ;-) > > On a side note: we configured correct weight per drive as well. > Ceph will neve

Re: [ceph-users] Unbalanced OSD's

2016-12-30 Thread Kees Meijs
Hi Asley, We experience (using Hammer) a similar issue. Not that I have a perfect solution to share, but I felt like mentioning a "me too". ;-) On a side note: we configured correct weight per drive as well. Regards, Kees On 29-12-16 11:54, Ashley Merrick wrote: > > Hello, > > > > I currently

Re: [ceph-users] How to know if an object is stored in clients?

2016-12-30 Thread Jaemyoun Lee
Thanks a lot! On Fri, Dec 30, 2016 at 6:32 PM Shinobu Kinjo wrote: > You can track activity of acting set by using: > > # ceph daemon osd.${osd id} dump_ops_in_flight > > On Fri, Dec 30, 2016 at 3:59 PM, Jaemyoun Lee > wrote: > > Dear Wido, > Is there a command to check the ACK? Or, may you te

Re: [ceph-users] How to know if an object is stored in clients?

2016-12-30 Thread Shinobu Kinjo
You can track activity of acting set by using: # ceph daemon osd.${osd id} dump_ops_in_flight On Fri, Dec 30, 2016 at 3:59 PM, Jaemyoun Lee wrote: > Dear Wido, > Is there a command to check the ACK? Or, may you tell me a source code > function for the received ACK? > > Thanks, > Jae > > On Thu

[ceph-users] installation docs

2016-12-30 Thread Manuel Sopena Ballesteros
Hi, I just would like to point a couple of issues I have following the INSTALLATION (QUICK) document. 1. The order to clean ceph deployment is: a. Ceph-deploy purge {ceph-node} [{ceph-node}] b. Ceph-deploy purgedata {ceph-node} [{ceph-node}] 2. I run ceph jewel 10.2.