[Gluster-users] Gluster 3.3.0 doesn't see neighbour

[Philip Poten]

Hi Vladimir

To circumvent any remaining /etc/hosts abnormalities, try with IPs
only. if that doesn't work either, you have a problem outside name
resolution. If it works with IPs, the problem is the resolver.

Note that nscd caches answers, also the "host" command uses DNS
responses, /not/ the /etc/hosts (nsswitch.conf order resp.) file - but
gluster does, as does "ping" and anything else using gethostbyname(3)

hth,
Philip

2012/7/2 Vladimir Yakovlev <mrquesty at gmail.com>:
> Thanks, Brian, for the hint.
> I've changed /etc/hosts with respect to your comment, but it didn't help
> either.
>
> The problem (from my perspective) in smth else, e.g. why, when I try to do
> the following, I see the blank response in tcpdump:
> [root at host1 ~]# ip a
> 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
>     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
>     inet 127.0.0.1/8 scope host lo
> 2: eth0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast state
> DOWN qlen 1000
>     link/ether 00:25:90:30:34:42 brd ff:ff:ff:ff:ff:ff
>     inet 46.../27 brd 46.182.25.159 scope global eth0
> 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state
> UP qlen 1000
>     link/ether 00:25:90:30:34:43 brd ff:ff:ff:ff:ff:ff
>     inet 192.168.1.192/24 brd 192.168.1.255 scope global eth1
> [root at host1 ~]# tcpdump -i eth1 'port 27004'
> tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
> listening on eth1, link-type EN10MB (Ethernet), capture size 65535 bytes
> ^Z
> [2]+  Stopped                 tcpdump -i eth1 'port 27004'
> [root at host1 ~]# bg
> [2]+ tcpdump -i eth1 'port 27004' &
>
> [root at host1 ~]# gluster peer probe host2
> Probe on localhost not needed
> [root at host1 ~]# fg
> tcpdump -i eth1 'port 27004'
> ^C
> 0 packets captured
> 117 packets received by filter
> 52 packets dropped by kernel
>
> So (by whatever reason) Gluster doesn't send a packet to anywhere through
> ethernet.
>
> Guys, any ideas?
>
> Thanks,
> BR,
> vy
>
>
> On Mon, Jul 2, 2012 at 6:06 PM, Brian Candler <B.Candler at pobox.com> wrote:
>>
>> On Mon, Jul 02, 2012 at 05:54:46PM +0400, Vladimir Yakovlev wrote:
>> >    I tried different configurations, the latest follows:
>> >    [root at host1 ~]# cat /etc/hosts
>> >    127.0.0.1   localhost localhost.localdomain localhost4
>> >    localhost4.localdomain4
>> >    127.0.0.1   host1 host1.fqdn
>> >    192.168.1.193   host2 host2.fqdn
>> >    [root at host2 ~]# cat /etc/hosts
>> >    127.0.0.1   localhost localhost.localdomain localhost4
>> >    localhost4.localdomain4
>> >    127.0.0.1   host2 host2.fqdn
>> >    192.168.1.192   host1 host1.fqdn
>>
>> Be consistent. Both machines should have:
>>
>> 127.0.0.1      localhost localhost.localdomain
>> 192.168.1.192  host1.fqdn host1
>> 192.168.1.193  host2.fqdn host2
>>
>> On host1 you should have "host1.fqdn" in /etc/hostname, and the command
>> "hostname" should show it. (Ditto for host2, but with "host2.fqdn" of
>> course)
>
>
>
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
>