[ceph-users] ceph.conf mon_max_pg_per_osd not recognized / set

Steven Vacaroaia stef97 at gmail.com
Wed Oct 31 07:16:59 PDT 2018


so, moving the entry from [mon] to [global] worked
This is a bit confusing - I use to put all my configuration setting
starting with mon_ under [mon]

Steven

On Wed, 31 Oct 2018 at 10:13, Steven Vacaroaia <stef97 at gmail.com> wrote:

> I do not think so ..or maybe I did not understand what are you saying
> There is no key listed on mgr config
>
> ceph config-key list
> [
>     "config-history/1/",
>     "config-history/2/",
>     "config-history/2/+mgr/mgr/dashboard/server_addr",
>     "config-history/3/",
>     "config-history/3/+mgr/mgr/prometheus/server_addr",
>     "config-history/4/",
>     "config-history/4/+mgr/mgr/dashboard/username",
>     "config-history/5/",
>     "config-history/5/+mgr/mgr/dashboard/password",
>     "config-history/6/",
>     "config-history/6/+mgr/mgr/balancer/mode",
>     "config-history/7/",
>     "config-history/7/+mgr/mgr/balancer/active",
>     "config-history/8/",
>     "config-history/8/+mgr/mgr/dashboard/password",
>     "config-history/8/-mgr/mgr/dashboard/password",
>     "config/mgr/mgr/balancer/active",
>     "config/mgr/mgr/balancer/mode",
>     "config/mgr/mgr/dashboard/password",
>     "config/mgr/mgr/dashboard/server_addr",
>     "config/mgr/mgr/dashboard/username",
>     "config/mgr/mgr/prometheus/server_addr",
>     "mgr/dashboard/crt",
>     "mgr/dashboard/key"
>
>
> On Wed, 31 Oct 2018 at 09:59, <ceph at jack.fr.eu.org> wrote:
>
>> Isn't this a mgr variable ?
>>
>> On 10/31/2018 02:49 PM, Steven Vacaroaia wrote:
>> > Hi,
>> >
>> > Any idea why different value for  mon_max_pg_per_osd is not
>> "recognized" ?
>> > I am using mimic 13.2.2
>> >
>> > Here is what I have in /etc/ceph/ceph.conf
>> >
>> >
>> > [mon]
>> > mon_allow_pool_delete = true
>> > mon_osd_min_down_reporters = 1
>> > mon_max_pg_per_osd = 400
>> >
>> > checking the value with
>> > ceph daemon osd.6 config show| grep mon_max_pg_per_osd still shows the
>> > default ( 250)
>> >
>> >
>> > Injecting a different value appears to works
>> > ceph tell osd.* injectargs '--mon_max_pg_per_osd 500'
>> >
>> > ceph daemon osd.6 config show| grep mon_max_pg_per_osd
>> >     "mon_max_pg_per_osd": "500",
>> >
>> > BUT
>> >
>> > cluster is still complaining TOO_MANY_PGS too many PGs per OSD (262 >
>> > max 250)
>> >
>> > I have restarted ceph.target services on monitor/manager server
>> > What else has to be done to have the cluster using the new value ?
>> >
>> > Steven
>> >
>> >
>> >
>> > _______________________________________________
>> > ceph-users mailing list
>> > ceph-users at lists.ceph.com
>> > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
>> >
>> _______________________________________________
>> ceph-users mailing list
>> ceph-users at lists.ceph.com
>> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ceph.com/pipermail/ceph-users-ceph.com/attachments/20181031/39c295dd/attachment.html>


More information about the ceph-users mailing list