[ceph-users] node not using cluster subnet

Steven Vacaroaia stef97 at gmail.com
Tue Oct 30 11:24:39 PDT 2018


Thanks for taking the trouble to provide advice
I found that the Juniper switch port for the server that did not work did
not have the MTU changed to 9200
I am using MTU 9000 for the cluster network

Not sure why packet fragmentation created issues but ...all seems fine now

Thanks
Steven

I


On Tue, 30 Oct 2018 at 13:22, Gregory Farnum <gfarnum at redhat.com> wrote:

> The OSDs ping each other on both public and cluster networks. Perhaps the
> routing isn't working on the public network? Or maybe it's trying to ping
> from the cluster 192. network into the public 10. network and that isn't
> getting through?
> -Greg
>
> On Tue, Oct 30, 2018 at 8:34 AM Steven Vacaroaia <stef97 at gmail.com> wrote:
>
>> Hi,
>> I am trying to add another node to my cluster which is configured to use
>> a dedicated subnet
>>
>> public_network = 10.10.35.0/24
>> cluster_network = 192.168.200.0/24
>>
>> For whatever reason, this node is staring properly and few seconds later
>> is failing
>> and staring to check for connectivity on public network
>>
>> The other 3 nodes are working fine
>> Nodes are identical
>>
>> Using kernel 4.18 and Mimic 13.2.2
>>
>> No firewall is involved
>>
>> I am really puzzled by this - any suggestions will be appreciated
>>
>> I have purged and reinstalled - also make sure I can ping using cluster
>> network
>>
>> 2018-10-30 11:09:28.344 7f274b537700  1 osd.3 308 state: booting -> active
>> 2018-10-30 11:09:29.621 7f275b848700  0 -- 192.168.200.204:6800/18679 >>
>> 192.168.200.201:6802/5008172 conn(0x557ed0318600 :6800
>> s=STATE_ACCEPTING_WAIT_CONNECT_MSG_AUTH pgs=0 cs=0 l=0).handle_connect_msg:
>> challenging authorizer
>> 2018-10-30 11:09:29.621 7f275b047700  0 -- 192.168.200.204:6800/18679 >>
>> 192.168.200.203:6800/6002192 conn(0x557ed0318c00 :6800
>> s=STATE_ACCEPTING_WAIT_CONNECT_MSG_AUTH pgs=0 cs=0 l=0).handle_connect_msg:
>> challenging authorizer
>> 2018-10-30 11:09:29.621 7f275b848700  0 -- 192.168.200.204:6800/18679 >>
>> 192.168.200.201:6802/5008172 conn(0x557ed0318000 :-1
>> s=STATE_ACCEPTING_WAIT_CONNECT_MSG_AUTH pgs=0 cs=0 l=0).handle_connect_msg:
>> challenging authorizer
>> 2018-10-30 11:09:29.621 7f275b047700  0 -- 192.168.200.204:6800/18679 >>
>> 192.168.200.203:6800/6002192 conn(0x557ed0319800 :-1
>> s=STATE_ACCEPTING_WAIT_CONNECT_MSG_AUTH pgs=0 cs=0 l=0).handle_connect_msg:
>> challenging authorizer
>> 2018-10-30 11:09:49.923 7f2756d4e700 -1 osd.3 308 heartbeat_check: no
>> reply from 10.10.35.201:6802 osd.0 ever on either front or back, first
>> ping sent 2018-10-30 11:09:29.621624 (cutoff 2018-10-30 11:09:29.924534)
>> 2018-10-30 11:09:49.923 7f2756d4e700 -1 osd.3 308 heartbeat_check: no
>> reply from 10.10.35.202:6802 osd.1 ever on either front or back, first
>> ping sent 2018-10-30 11:09:29.621624 (cutoff 2018-10-30 11:09:29.924534)
>>
>> _______________________________________________
>> 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/20181030/e0d1d4f8/attachment.html>


More information about the ceph-users mailing list