Hi everyone,
I inherited a large Ganglia 2.0 installation which I am currently
trying to upgrade to version 2.5.4. I decided to first roll this out to a
new ~200 node Linux cluster which has never had Ganglia in hopes of
familiarizing myself with this tool. I installed the gmond RPM on all the
nodes. I didn't touch the /etc/gmond.conf file at all.
Once this was done, I telnet to port 8649 on the localhost and
received the XML dump that I expected...all the hosts were at least listed
in there. However, when I run gstat, it sees all the nodes as being dead,
and my log files are filling up very fast with stuff like this:
Aug 26 15:32:07 l-sim-205-145 /usr/sbin/gmond[683]: mcast_listen_thread()
error: STRANGE type!
Aug 26 15:32:07 l-sim-205-145 /usr/sbin/gmond[683]: mcast_listen_thread()
error: STRANGE type!
Aug 26 15:32:07 l-sim-205-145 /usr/sbin/gmond[685]: mcast_listen_thread()
xdr_string() error: Interrupted system call
Aug 26 15:32:07 l-sim-205-145 /usr/sbin/gmond[683]: mcast_listen_thread()
error: STRANGE type!
Aug 26 15:32:08 l-sim-205-145 /usr/sbin/gmond[683]: mcast_listen_thread()
xdr_string() error: Interrupted system call
Aug 26 15:32:09 l-sim-205-145 /usr/sbin/gmond[685]: mcast_listen_thread()
xdr_string() error: Interrupted system call
Aug 26 15:32:09 l-sim-205-145 /usr/sbin/gmond[683]: mcast_listen_thread()
xdr_string() error: Interrupted system call
Aug 26 15:32:09 l-sim-205-145 /usr/sbin/gmond[683]: mcast_listen_thread()
xdr_string() error: Interrupted system call
Aug 26 15:32:10 l-sim-205-145 /usr/sbin/gmond[685]: mcast_listen_thread()
xdr_string() error: Interrupted system call
Aug 26 15:32:11 l-sim-205-145 /usr/sbin/gmond[685]: mcast_listen_thread()
error: STRANGE type!
Aug 26 15:32:11 l-sim-205-145 /usr/sbin/gmond[685]: mcast_listen_thread()
error: STRANGE type!
Aug 26 15:32:11 l-sim-205-145 /usr/sbin/gmond[685]: mcast_listen_thread()
xdr_string() error: Interrupted system call
Aug 26 15:32:11 l-sim-205-145 /usr/sbin/gmond[685]: mcast_listen_thread()
xdr_string() error: Interrupted system call
Aug 26 15:32:13 l-sim-205-145 /usr/sbin/gmond[683]: mcast_listen_thread()
xdr_string() error: Interrupted system call
Aug 26 15:32:13 l-sim-205-145 /usr/sbin/gmond[683]: mcast_listen_thread()
xdr_string() error: Interrupted system call
Aug 26 15:32:13 l-sim-205-145 /usr/sbin/gmond[685]: mcast_listen_thread()
error: STRANGE type!
Aug 26 15:32:13 l-sim-205-145 /usr/sbin/gmond[685]: mcast_listen_thread()
xdr_string() error: Interrupted system call
Aug 26 15:32:13 l-sim-205-145 /usr/sbin/gmond[685]: mcast_listen_thread()
xdr_string() error: Interrupted system call
Aug 26 15:32:13 l-sim-205-145 /usr/sbin/gmond[683]: mcast_listen_thread()
error: STRANGE type!
Aug 26 15:32:14 l-sim-205-145 /usr/sbin/gmond[683]: mcast_listen_thread()
xdr_string() error: Interrupted system call
Aug 26 15:32:15 l-sim-205-145 /usr/sbin/gmond[685]: mcast_listen_thread()
error: STRANGE type!
...any ideas what I'm doing wrong? I'm not very familiar at all with
multicast. Thanks a lot for any help.
Steve Gilbert
Unix Systems Administrator
[EMAIL PROTECTED]
