Hi Gowrishankar,

 

this problem seems to be related to cluster, not to bonding: bonding is working correctly, anyway I’ve tryied a test removing bonding, and I experience the same problem directly on interface eth0.

 

This is my cluster.conf

 

<?xml version="1.0" ?>

<cluster alias="cluster01" config_version="54" name="cluster01">

        <fence_daemon clean_start="1" post_fail_delay="0" post_join_delay="30"/>

        <clusternodes>

                <clusternode name="AREA041" nodeid="2" votes="1">

                        <fence/>

                </clusternode>

                <clusternode name="AREA042" nodeid="3" votes="1">

                        <fence/>

                </clusternode>

        </clusternodes>

        <cman expected_votes="1" two_node="1"/>

        <fencedevices/>

        <rm>

                <failoverdomains>

                        <failoverdomain name="httpd failover domain" ordered="0" restricted="1">

                                <failoverdomainnode name="AREA041" priority="1"/>

                        </failoverdomain>

                </failoverdomains>

                <resources>

                        <ip address="10.0.181.3" monitor_link="1"/>

                </resources>

                <service autostart="0" domain="httpd failover domain" name="Apache" recovery="disable">

                        <script file="/etc/rc.d/init.d/httpd" name="script httpd"/>

                        <ip ref="10.0.181.3"/>

                </service>

                <service autostart="0" domain="httpd failover domain" name="Service Mail" recovery="disable">

                        <script file="/etc/rc.d/init.d/MailScanner" name="MailScanner"/>

                        <clusterfs device="/dev/DATI_MAIL/DATI_MAIL" force_unmount="1" fsid="5845" fstype="gfs2" mountpoint="/dati_mail" name="Share_dati_mail" options=""/>

                        <ip address="10.0.181.4" monitor_link="1"/>

                </service>

        </rm>

</cluster>

 

Many thanks

 

Fabio

 

 

 

From: centos-bounces@centos.org [mailto:centos-bounces@centos.org] On Behalf Of A Linux Fanatic
Sent: martedì 23 dicembre 2008 11:43
To: CentOS mailing list
Subject: Re: [CentOS] cluster - ip address lost when service stopped

 

Fabio Macchi wrote: 

 
-----Original Message-----
From: centos-bounces@centos.org [mailto:centos-bounces@centos.org] On Behalf Of A Linux Fanatic
Sent: martedì 23 dicembre 2008 06:08
To: CentOS mailing list
Subject: Re: [CentOS] cluster - ip address lost when service stopped
 
Fabio Macchi wrote:
  
Hi all,
 
I'm trying to setup a cluster of 2 machines with Centos 5.2 to host a 
postfix+spamassassin+clamav+mailscanner service. Below cluster 
software versions
 
rgmanager.i386 2.0.38-2.el5_2.1 installed
 
cman.i386 2.0.84-2.el5_2.2 installed
 
Every machine (hp blade server ) has 4 interfaces, bounded in this way:
 
Eth0, eth1 -> bond0 -> connection for public service ( 10.0.181.x )
 
Eth2,eth3 -> bond1 -> connection for intra-cluster communication ( 
192.168.44.x )
 
bond0 Link encap:Ethernet HWaddr 00:21:5A:48:DA:BE
 
inet addr:10.0.181.41 Bcast:10.0.181.255 Mask:255.255.255.0
 
inet6 addr: fe80::221:5aff:fe48:dabe/64 Scope:Link
 
UP BROADCAST RUNNING MASTER MULTICAST MTU:1500 Metric:1
 
RX packets:85 errors:0 dropped:0 overruns:0 frame:0
 
TX packets:86 errors:0 dropped:0 overruns:0 carrier:0
 
collisions:0 txqueuelen:0
 
RX bytes:12963 (12.6 KiB) TX bytes:9144 (8.9 KiB)
 
bond1 Link encap:Ethernet HWaddr 00:1F:29:6D:7D:08
 
inet addr:192.168.44.41 Bcast:192.168.44.255 Mask:255.255.255.0
 
inet6 addr: fe80::21f:29ff:fe6d:7d08/64 Scope:Link
 
UP BROADCAST RUNNING MASTER MULTICAST MTU:1500 Metric:1
 
RX packets:29 errors:0 dropped:0 overruns:0 frame:0
 
TX packets:223 errors:0 dropped:0 overruns:0 carrier:0
 
collisions:0 txqueuelen:0
 
RX bytes:4612 (4.5 KiB) TX bytes:31746 (31.0 KiB)
 
Then I've created a new Mail service with these local resources:
 
- Ip address 10.0.181.3
 
- Script /etc/rc.d/init.d/MailScanner
 
- GFS file system on a SAN
 
Service start, but the problem is that, when I stop the service, 
external ip address is removed from bond0.
 
 
    
 
Hi Fabio,
 
Could you please attach the following files:
 
/etc/sysconfig/network-scripts/ifcfg-bond0
/etc/sysconfig/network-scripts/ifcfg-bond1
/etc/sysconfig/network-scripts/ifcfg-eth0
/etc/sysconfig/network-scripts/ifcfg-eth1
/etc/sysconfig/network-scripts/ifcfg-eth2
/etc/sysconfig/network-scripts/ifcfg-eth3
/etc/cluster/cluster.conf
 
And "external ip address is removed from bond0." - I assume here 
external IP is 10.0.181.41, right?
 
Thanks
Gowrishankar Rajaiyan | A Linux Fanatic.
_______________________________________________
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
 
 
 
Hi Gowrishankar,
 
requested files attached; you well understand, I mean ip 10.0.181.41 disappear ( below the output from ifconfig after I tried to stop the service)
 
 
bond0     Link encap:Ethernet  HWaddr 00:21:5A:48:DA:BE  
          inet6 addr: fe80::221:5aff:fe48:dabe/64 Scope:Link
          UP BROADCAST RUNNING MASTER MULTICAST  MTU:1500  Metric:1
          RX packets:52958 errors:0 dropped:0 overruns:0 frame:0
          TX packets:7844 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0 
          RX bytes:4915061 (4.6 MiB)  TX bytes:4936239 (4.7 MiB)
 
Tks
 
Fabio
 


Hi Fabio


First, issue the command:

 $ rpm -qf /sbin/ifup

It will respond with a line of text starting with either "initscripts" or "sysconfig," followed by some numbers. This is the package that provides your network initialization scripts.

Next, to determine if your installation supports bonding, issue the command:

 $ grep ifenslave /sbin/ifup

If this returns any matches, then your initscripts or sysconfig has support for bonding.


Ref: http://www.linuxfoundation.org/en/Net:Bonding

Try configuring ifcfg-bondX using the contents described in the above link.

Thanks
Rajaiyan | A Linux Fanatic.