Re: [Gluster-users] Does gluster passes LAN boundaries?
Amar Tumballi writes: > > > > > > > Could you point to me were this log is located?? On the servers I have a > > /var/log/glusterfs folder but in the client I cannot see where the logs are > > located.. > > > > > Do 'which glusterfs' and get the prefix. The logs will be located by default > @ '$prefix/var/log/glusterfs/*' > > Regards, > Amar > > > ___ > Gluster-users mailing list > gluster-us...@... > http://gluster.org/cgi-bin/mailman/listinfo/gluster-users > Have you figured this issue out yet? I am experiencing the same issue with my 4 node cluster. I have 4 PowerEdge 2650's running hardware RAID 0 with CentOS 5.5 installed, Linux Kernel 2.6.18-194.26.1.el5xen, Gluster 3.1.1 , and where all four servers (192.168.1.201-204) contain a brick volume for the Gluster. I also have one of the four servers configured as a client (.201) in addition to serving one of the brick volumes. When I issue a df -h command it hangs until I kill the PID or reboot. Nothing ever displays as output, nothing in log files so far, and I can’t write to the mount point folder. By the way, I can successfully issue the mount command with success, it until after that which the problem starts to occur. Dem ___ Gluster-users mailing list Gluster-users@gluster.org http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
Re: [Gluster-users] Does gluster passes LAN boundaries?
> > > Could you point to me were this log is located?? On the servers I have a > /var/log/glusterfs folder but in the client I cannot see where the logs are > located.. > > Do 'which glusterfs' and get the prefix. The logs will be located by default @ '$prefix/var/log/glusterfs/*' Regards, Amar ___ Gluster-users mailing list Gluster-users@gluster.org http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
Re: [Gluster-users] Does gluster passes LAN boundaries?
It may be firewalls preventing port 24007 and above.. By default most of the firewall rules allow ssh/nfs/ftp etc. Check by flushing the iptables rules to be sure. Also, Check the log file on the client machine which may give some hints. Regards, Amar On Wed, Dec 1, 2010 at 12:08 PM, Horacio Sanson wrote: > On Wednesday 01 December 2010 15:15:49 Horacio Sanson wrote: > > I have four gluster nodes with several volumes configured that work > > perfectly with all my clients (64bit and 32bit) on the LAN but I have a > > client in a different IP subnetwork that has trouble accessing the > > volumes. > > > > I can mount the gluster volumes but issuing ls or df commands on the > mount > > point hangs forever. > > > > > > This is my configuration: > > > > 4 Gluster Server Nodes: > >IP range: 192.168.4.90 to 192.168.4.93 > >OS: Ubuntu 10.10 64bit > > Kernel: 2.6.35-22-server > > GlusterFS: 3.1.1-stable installed from deb package > > > >One client that works without problems > >IP: 192.168.4.111 > >OS: Ubuntu 10.10 32bit > > Kernel: 2.6.35-22-generic > > Gluster: 3.1.1-stable installed from source > > > >One client that does not work!! > > IP: 192.168.0.228 > > OS: Unbuntu 8.10 32 bit > > Kernel: 2.6.24-28-generic > > Gluster: 3.1.1-stable installed from source > > Problem: Can mount volumes but ls and df hang forever on the mount > > points. > > > > > > The differences between the client that works and the one that does not > > work are that they are in different networks and the kernel version. I > can > > access via ssh/ping/etc to the problem client so it is not a networking > > problem. > > Forgot to mention that I can mount the volumes using nfs without problems > from > the problematic client. > > -- > regards, > Horacio Sanson > ___ > Gluster-users mailing list > Gluster-users@gluster.org > http://gluster.org/cgi-bin/mailman/listinfo/gluster-users > ___ Gluster-users mailing list Gluster-users@gluster.org http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
