[Deepsea-users] "no ceph clusters available"

[Robert Grosschopff]

Alejandro,

the Crushmap still doesn't show up after fixing the warning about having too many PGs. That warning appeared  right after SES was installed because some initial pools are created with a standard number of PGs unrelated to the number of OSDs. It worked at that time. 
After rebooting the cluster the Nodes tab stays empty (apart from the title “Ceph Hosts”)  for whatever reason. 

Robert

On 15/03/2017, 16:41, "deepsea-users-bounces at lists.suse.com on behalf of Alejandro Bonilla" <deepsea-users-bounces at lists.suse.com on behalf of abonilla at suse.com> wrote:

    Robert,
    
    You need to fix the Warning and then it won’t give out the confusing error msg...
    
    > On Mar 15, 2017, at 10:40 AM, Robert Grosschopff <Robert.Grosschopff at suse.com> wrote:
    > 
    > Hi Alejandro,
    > 
    > adding widgets works like a charm (
    > 
    > All other files in /etc/ceph are present as well.
    > 
    > I can also extract the crushmap using ceph osd getcrushmap –o cmap.bin
    > 
    > Weird, somehow
    > 
    > Robert
    > 
    > 
    > On 15/03/2017, 15:10, "deepsea-users-bounces at lists.suse.com on behalf of Alejandro Bonilla" <deepsea-users-bounces at lists.suse.com on behalf of abonilla at suse.com> wrote:
    > 
    >    Robert,
    > 
    >    I think that “error” is normal. Just try adding widgets and see if the cluster shows up on the field as “ceph”…
    > 
    >    Also, fix the WARNS too… If that doesn’t do it make sure of:
    > 
    >    ls -l /etc/ceph/
    >    total 16
    >    -rw------- 1 root      root      129 Jan 24 23:15 ceph.client.admin.keyring
    >    -rw-rw---- 1 openattic openattic 111 Jan 24 23:17 ceph.client.openattic.keyring
    >    -rw-r--r-- 1 root      root      362 Jan 24 23:15 ceph.conf
    > 
    >    DeepSea should have done that.
    > 
    >> On Mar 15, 2017, at 8:33 AM, Robert Grosschopff <Robert.Grosschopff at suse.com> wrote:
    >> 
    >> Yepp, apart from the usual “too many PGs per OSD” everything is fine on the command line.
    >> 
    >> openATTIC also shows all OSDs, Pools, Nodes, etc.
    >> 
    >> 
    >> 
    >> On 15/03/2017, 13:30, "deepsea-users-bounces at lists.suse.com on behalf of Lars Marowsky-Bree" <deepsea-users-bounces at lists.suse.com on behalf of lmb at suse.com> wrote:
    >> 
    >>   On 2017-03-15T12:28:02, Robert Grosschopff <Robert.Grosschopff at suse.com> wrote:
    >> 
    >>> I recently installed a cluster using Salt. Everything was fine and I could see a graphical representation of the Crushmap in openATTIC.
    >>> 
    >>> After a reboot of the cluster I keep getting a “Sorra folks … no ceph clusters available" .
    >> 
    >>   Is the cluster itself running? 
    >> 
    >> 
    >>   _______________________________________________
    >>   Deepsea-users mailing list
    >>   Deepsea-users at lists.suse.com
    >>   http://lists.suse.com/mailman/listinfo/deepsea-users
    >> 
    >> 
    >> _______________________________________________
    >> Deepsea-users mailing list
    >> Deepsea-users at lists.suse.com
    >> http://lists.suse.com/mailman/listinfo/deepsea-users
    > 
    >    _______________________________________________
    >    Deepsea-users mailing list
    >    Deepsea-users at lists.suse.com
    >    http://lists.suse.com/mailman/listinfo/deepsea-users
    > 
    > 
    > _______________________________________________
    > Deepsea-users mailing list
    > Deepsea-users at lists.suse.com
    > http://lists.suse.com/mailman/listinfo/deepsea-users
    
    _______________________________________________
    Deepsea-users mailing list
    Deepsea-users at lists.suse.com
    http://lists.suse.com/mailman/listinfo/deepsea-users