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

[John Mark Walker]

Awesome! Glad it worked out :) 

-JM 

----- Original Message -----

> Hi Everyone,

> Problem has been SOLVED.

> My colleague was playing with hearbeat on this machine and didn't
> forget to reset net.ipv4.ip_nonlocal_bind = 0. I fixed it, did
> sysctl -p and now:

> # gluster peer probe 192.168.1.193
> Probe successful

> Thanks, guys, for great support and sorry for keep you busy ;-)

> BR,
> Vladimir.

> On Mon, Jul 2, 2012 at 6:34 PM, Vladimir Yakovlev <
> mrquesty at gmail.com > wrote:

> > Hi Philip,
> 

> > I tried IPs instead of hostnames at the beginning, but it didn't
> > help:
> 

> > [root at host1 ~]# gluster peer probe 192.168.1.193
> 

> > Probe on localhost not needed
> 

> > My assumption is smth wrong smwhere else. Donno where - no ideas at
> > all; the worst is i have a set of test machines, where everything
> > works perfectly... :(
> 

> > There are two switches between these machines, but they allow
> > Gluster
> > traffic responsively.
> 

> > Hope, we can struggle this together ;-)
> 

> > Thanks, guys, for quick responses.
> 

> > Regards,
> 
> > vy
> 

> > On Mon, Jul 2, 2012 at 6:27 PM, Philip Poten <
> > philip.poten at gmail.com
> > > wrote:
> 

> > > 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
> > 
> 
> > > >
> > 
> 

> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://supercolony.gluster.org/pipermail/gluster-users/attachments/20120702/994d1687/attachment.html>