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://gluster.org/pipermail/gluster-users/attachments/20120702/994d1687/attachment-0001.htm>
