Hi all, I've spent the past little while attempting to setup Conga on a two node CentOS 6 cluster. While I do have the cluster up and running, I'm running into problems attempting to start services through luci on either node. A little background information: -Both nodes have dual NICs, one interface on each node will be used for the WAN, the other is a private gigabit crossover link between the nodes. -Both nodes are in /etc/hosts -The nodes are up to date, and both are running x86_64 -SELinux and iptables have both been disabled -To get the cluster running, luci, ricci, and their dependencies were installed, and 'service start ricci', then 'service start luci' where issued. ricci created certs when first run. The problem is: I've successfully created an IP resource through luci, but when attempting to start it through 'Services', it fails to start and I get the following error in /var/log/luci/luci.log: 07:32:33,648 INFO [luci.controllers.cluster] User "root" started service "ip failover" in cluster "MailCluster" 07:32:33,654 ERROR [luci.lib.ricci_communicator] Unable to establish an SSL connection to mail01.mydomain.com:11111: _ssl.c:480: error:140770FC:SSL routines:SSL23_GET_SERVER_HELLO:unknown protocol Traceback (most recent call last): File "/usr/lib/python2.6/site-packages/luci/lib/ricci_communicator.py", line 56, in __init__ self.ss.connect((self.__hostname, self.__port)) File "/usr/lib64/python2.6/ssl.py", line 309, in connect self.do_handshake() File "/usr/lib64/python2.6/ssl.py", line 293, in do_handshake self._sslobj.do_handshake() SSLError: [Errno 1] _ssl.c:480: error:140770FC:SSL routines:SSL23_GET_SERVER_HELLO:unknown protocol When I connect to port 11111 using openssl, I receive 'Clients_SSL_certificate_required' and the connection closes. Also, netstat reports: tcp 0 0 :::11111 :::* LISTEN 9745/ricci I hope I've provided enough information, but any input or thoughts would be greatly appreciated. Thanks, James
Hi James, I have been having all sort of issues playing with conga is 6.x as well. To the point that I am giving up on it until I find a how to properly configure this .>>> James Edwards <jedwards at bsdftw.org> 9/8/2011 9:36 AM >>>Hi all, I've spent the past little while attempting to setup Conga on a two node CentOS 6 cluster. While I do have the cluster up and running, I'm running into problems attempting to start services through luci on either node. A little background information: -Both nodes have dual NICs, one interface on each node will be used for the WAN, the other is a private gigabit crossover link between the nodes. -Both nodes are in /etc/hosts -The nodes are up to date, and both are running x86_64 -SELinux and iptables have both been disabled -To get the cluster running, luci, ricci, and their dependencies were installed, and 'service start ricci', then 'service start luci' where issued. ricci created certs when first run. The problem is: I've successfully created an IP resource through luci, but when attempting to start it through 'Services', it fails to start and I get the following error in /var/log/luci/luci.log: 07:32:33,648 INFO [luci.controllers.cluster] User "root" started service "ip failover" in cluster "MailCluster" 07:32:33,654 ERROR [luci.lib.ricci_communicator] Unable to establish an SSL connection to mail01.mydomain.com:11111: _ssl.c:480: error:140770FC:SSL routines:SSL23_GET_SERVER_HELLO:unknown protocol Traceback (most recent call last): File "/usr/lib/python2.6/site-packages/luci/lib/ricci_communicator.py", line 56, in __init__ self.ss.connect((self.__hostname, self.__port)) File "/usr/lib64/python2.6/ssl.py", line 309, in connect self.do_handshake() File "/usr/lib64/python2.6/ssl.py", line 293, in do_handshake self._sslobj.do_handshake() SSLError: [Errno 1] _ssl.c:480: error:140770FC:SSL routines:SSL23_GET_SERVER_HELLO:unknown protocol When I connect to port 11111 using openssl, I receive 'Clients_SSL_certificate_required' and the connection closes. Also, netstat reports: tcp 0 0 :::11111 :::* LISTEN 9745/ricci I hope I've provided enough information, but any input or thoughts would be greatly appreciated. Thanks, James _______________________________________________ CentOS mailing list CentOS at centos.org lists.centos.org/mailman/listinfo/centos -------------- next part -------------- An HTML attachment was scrubbed... URL: <lists.centos.org/pipermail/centos/attachments/20110908/127a546c/attachment.html>
On 9/8/2011 9:40 AM, Lisandro Grullon wrote:> Hi James, > I have been having all sort of issues playing with conga is 6.x as > well. To the point that I am giving up on it until I find a how to > properly configure this . >I've been pulling my hair out about this, and its just one of many things that has driven me nuts within 6.0. I was able to get this working in 5.6, but unfortunately, that is a route I can't take for this project because the powers that be are concerned about future support (i.e. updates). I personally just want what works and what will work. I've come across other somewhat viable solutions, but I'm also kind of limited in the options I can take, because without a web interface (or god forbid a GUI) I'm concerned that no one else would ever know what is going on with the systems/cluster.
Greetings, On Thu, Sep 8, 2011 at 7:06 PM, James Edwards <jedwards at bsdftw.org> wrote:> Hi all, > > I've spent the past little while attempting to setup Conga on a two node > CentOS 6 cluster. ?While I do have the cluster up and running, I'm > running into problems attempting to start services through luci on > either node.I sincerely feel that this query goes to linux-cluster list. You will mostly get answers there. -- Regards, Rajagopal