[Gluster-users] Ip based peer probe volume create error

[Ercan Aydoğan]

Yes, last node different subnet.

all servers on scaleway.com

node1 and node2 same subnet
node3 another subnet

But all can ping each other.

how about uuid generation ?

i looked code and if i check correct path it use hostname for generating uuid.

Maybe peerinfo->hostname is not same and rejecting peer with failed: Host 51.15.90.60 is not in 'Peer in Cluster' state 

root at pri:~# ping 163.172.151.120
PING 163.172.151.120 (163.172.151.120) 56(84) bytes of data.
64 bytes from 163.172.151.120: icmp_seq=1 ttl=52 time=13.1 ms
64 bytes from 163.172.151.120: icmp_seq=2 ttl=52 time=14.8 ms
64 bytes from 163.172.151.120: icmp_seq=3 ttl=52 time=16.8 ms
64 bytes from 163.172.151.120: icmp_seq=4 ttl=52 time=19.6 ms
64 bytes from 163.172.151.120: icmp_seq=5 ttl=52 time=19.8 ms
^C
--- 163.172.151.120 ping statistics ---
5 packets transmitted, 5 received, 0% packet loss, time 4006ms
rtt min/avg/max/mdev = 13.141/16.874/19.865/2.636 ms


root at sec:~# ping 51.15.77.14
PING 51.15.77.14 (51.15.77.14) 56(84) bytes of data.
64 bytes from 51.15.77.14: icmp_seq=1 ttl=59 time=5.66 ms
64 bytes from 51.15.77.14: icmp_seq=2 ttl=59 time=1.44 ms
64 bytes from 51.15.77.14: icmp_seq=3 ttl=59 time=1.39 ms
64 bytes from 51.15.77.14: icmp_seq=4 ttl=59 time=0.769 ms
64 bytes from 51.15.77.14: icmp_seq=5 ttl=59 time=5.05 ms
64 bytes from 51.15.77.14: icmp_seq=6 ttl=59 time=7.08 ms
64 bytes from 51.15.77.14: icmp_seq=7 ttl=59 time=7.50 ms
64 bytes from 51.15.77.14: icmp_seq=8 ttl=59 time=5.04 ms
^C
--- 51.15.77.14 ping statistics ---
8 packets transmitted, 8 received, 0% packet loss, time 7007ms
rtt min/avg/max/mdev = 0.769/4.246/7.509/2.502 ms

root at third:~# ping 51.15.77.14
PING 51.15.77.14 (51.15.77.14) 56(84) bytes of data.
64 bytes from 51.15.77.14: icmp_seq=1 ttl=51 time=20.3 ms
64 bytes from 51.15.77.14: icmp_seq=2 ttl=51 time=21.5 ms
64 bytes from 51.15.77.14: icmp_seq=3 ttl=51 time=20.3 ms
64 bytes from 51.15.77.14: icmp_seq=4 ttl=51 time=21.4 ms
64 bytes from 51.15.77.14: icmp_seq=5 ttl=51 time=21.9 ms
64 bytes from 51.15.77.14: icmp_seq=6 ttl=51 time=18.1 ms
^C
--- 51.15.77.14 ping statistics ---
6 packets transmitted, 6 received, 0% packet loss, time 5006ms
rtt min/avg/max/mdev = 18.151/20.629/21.942/1.267 ms

> On 8 Feb 2018, at 01:13, Lindsay Mathieson <lindsay.mathieson at gmail.com> wrote:
> 
> On 8/02/2018 4:45 AM, Gaurav Yadav wrote:
>> After seeing command history, I could see that you have 3 nodes, and firstly you are peer probing 51.15.90.60  and 163.172.151.120 from  51.15.77.14
>> So here itself you have 3 node cluster, after all this you are going on node 2 and again peer probing 51.15.77.14.
>>  Ideally it should work, with above steps, but due to some address resolution volume create is getting failed, and from glusterd logs its not clear.
> 
> Given 163.172.151.120 is on a different subnet I'd guess routing issues. Can it be pinged from both the other nodes?
> 
> -- 
> Lindsay
> 
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://lists.gluster.org/mailman/listinfo/gluster-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20180208/52d30f88/attachment.html>