To see where your OSDs and Mon are listening, you have various cmds in Linux, e.g:
'lsof -ni | grep ceph' - you should see one LISTEN line for the monitor, 2 LISTEN lines for the OSDs and a lot of ESTABLISHED lines, which indicate communication between OSDs and OSDs and clients 'netstat -atn | grep LIST' - you should see a lot of lines with portnumber 6800 and upwards (OSDs) and port 6789 (MON) More comments inline. HTH, Kurt > Gandalf Corvotempesta <mailto:gandalf.corvotempe...@gmail.com> > 28. April 2014 11:05 > 2014-04-26 12:06 GMT+02:00 Gandalf Corvotempesta > > I've added "cluster addr" and "public addr" to each OSD configuration > but nothing is changed. > I see all OSDs down except the ones from one server but I'm able to > ping each other nodes on both interfaces. What do you mean by "I see all OSDs down"? What does a 'ceph osd stat' say? > > How can I detect what ceph is doing? 'ceph -w' > I see tons of debug logs but they > are not very easy to understand > with "ceph health" i can see that "pgs down" value is slowly > decreasing so I can suppose that caph is recovering. Is that right? What's the output of 'ceph -s' > > Isn't possible to add a semplified output like the one coming from > "mdadm"? (cat /proc/mdstat) > _______________________________________________ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > Gandalf Corvotempesta <mailto:gandalf.corvotempe...@gmail.com> > 26. April 2014 12:06 > I've not defined cluster IPs for each OSD server but only the whole > subnet. > Should I define each IP for each OSD ? This is not wrote on docs and > could be tricky to do this in big environments with hundreds of nodes > _______________________________________________ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > McNamara, Bradley <mailto:bradley.mcnam...@seattle.gov> > 24. April 2014 20:04 > Do you have all of the cluster IP's defined in the host file on each > OSD server? As I understand it, the mon's do not use a cluster > network, only the OSD servers. > > -----Original Message----- > From: ceph-users-boun...@lists.ceph.com > [mailto:ceph-users-boun...@lists.ceph.com] On Behalf Of Gandalf > Corvotempesta > Sent: Thursday, April 24, 2014 8:54 AM > To: ceph-users@lists.ceph.com > Subject: [ceph-users] cluster_network ignored > > I'm trying to configure a small ceph cluster with both public and > cluster networks. > This is my conf: > > [global] > public_network = 192.168.0/24 > cluster_network = 10.0.0.0/24 > auth cluster required = cephx > auth service required = cephx > auth client required = cephx > fsid = 004baba0-74dc-4429-84ec-1e376fb7bcad > osd pool default pg num = 8192 > osd pool default pgp num = 8192 > osd pool default size = 3 > > [mon] > mon osd down out interval = 600 > mon osd mon down reporters = 7 > [mon.osd1] > host = osd1 > mon addr = 192.168.0.1 > [mon.osd2] > host = osd2 > mon addr = 192.168.0.2 > [mon.osd3] > host = osd3 > mon addr = 192.168.0.3 > > [osd] > osd mkfs type = xfs > osd journal size = 16384 > osd mon heartbeat interval = 30 > filestore merge threshold = 40 > filestore split multiple = 8 > osd op threads = 8 > osd recovery max active = 5 > osd max backfills = 2 > osd recovery op priority = 2 > > > on each node I have bond0 bound to 192.168.0.x and bond1 bound to > 10.0.0.x When ceph is doing recovery, I can see replication through > bond0 (public interface) and nothing via bond1 (cluster interface) > > Should I configure anything else ? > _______________________________________________ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > > > _______________________________________________ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > Gandalf Corvotempesta <mailto:gandalf.corvotempe...@gmail.com> > 24. April 2014 17:53 > I'm trying to configure a small ceph cluster with both public and > cluster networks. > This is my conf: > > [global] > public_network = 192.168.0/24 > cluster_network = 10.0.0.0/24 > auth cluster required = cephx > auth service required = cephx > auth client required = cephx > fsid = 004baba0-74dc-4429-84ec-1e376fb7bcad > osd pool default pg num = 8192 > osd pool default pgp num = 8192 > osd pool default size = 3 > > [mon] > mon osd down out interval = 600 > mon osd mon down reporters = 7 > [mon.osd1] > host = osd1 > mon addr = 192.168.0.1 > [mon.osd2] > host = osd2 > mon addr = 192.168.0.2 > [mon.osd3] > host = osd3 > mon addr = 192.168.0.3 > > [osd] > osd mkfs type = xfs > osd journal size = 16384 > osd mon heartbeat interval = 30 > filestore merge threshold = 40 > filestore split multiple = 8 > osd op threads = 8 > osd recovery max active = 5 > osd max backfills = 2 > osd recovery op priority = 2 > > > on each node I have bond0 bound to 192.168.0.x and bond1 bound to 10.0.0.x > When ceph is doing recovery, I can see replication through bond0 > (public interface) and nothing via bond1 (cluster interface) > > Should I configure anything else ? > _______________________________________________ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
smime.p7s
Description: S/MIME Cryptographic Signature
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
