[Linux-cluster] SOLVED Netmask bit set to 32 in service ip

Pavlos Parissis p_pavlos at freemail.gr
Fri Feb 15 08:13:47 UTC 2008


Quoting Lon Hohberger <lhh at redhat.com>:

> On Thu, 2008-02-14 at 02:59 +0200, Pavlos Parissis wrote:
>
>> ocsi2# patch -p0 <  /tmp/ip.sh-netmask.patch
>> patching file ip.sh
>
> That's strange - it worked great for me.  Could you run:
>
>    rg_test rules &> /tmp/foo.out
>
> Look at /tmp/foo.out for a line that says:
>
>    "Error storing ip: Duplicate"
>
> If you see that, try cleaning up /usr/share/cluster.  Look for files
> like:
>
>    /usr/share/cluster/ip.sh~
>    /usr/share/cluster/ip.rpm*
ocsi1# cd /usr/share/cluster/
ocsi1# ls -sl ip.*
20 -rwxr-xr-x  1 root root 18248 Aug  2  2006 ip.sh
ocsi1# cp -p ip.sh /root/ip.sh-ORIGINAL
ocsi1# grep -A 1 -B 4 '10.10.21.138' /etc/cluster/cluster.conf
     <service domain="FirstDomain" name="ppr1" autostart="1"  
recovery="relocate">
       <script name="ppr1" file="/usr/local/wsb/scripts/rhc_ppr1"/>
       <ip monitor_link="1" address="10.10.21.71"/>
       <fs device="/dev/sdd1" mountpoint="/usr/omg_ppr"  
force_unmount="1" name="/usr/omg_ppr" fstype="ext3"/>
       <ip monitor_link="1" address="10.10.21.138"/>
     </service>
ocsi1#  ip add sh|grep  -B 3 '10.10.21.138'
     link/ether 00:19:bb:3b:ac:89 brd ff:ff:ff:ff:ff:ff
     inet 10.10.21.132/26 brd 10.10.21.191 scope global bond0
     inet 10.10.21.142/32 scope global bond0
     inet 10.10.21.138/32 scope global bond0

ocsi1# cat /tmp/ip.sh-netmask.patch
Index: ip.sh
===================================================================
RCS file: /cvs/cluster/cluster/rgmanager/src/resources/ip.sh,v
retrieving revision 1.5.2.17
diff -u -r1.5.2.17 ip.sh
--- ip.sh       22 Mar 2007 23:15:59 -0000      1.5.2.17
+++ ip.sh       13 Feb 2008 18:28:12 -0000
@@ -692,6 +692,10 @@
                         if [ $? -ne 0 ]; then
                                 continue
                         fi
+
+                       if [ "${addr/\/*/}" = "${addr}" ]; then
+                               addr="$addr/$maskbits"
+                       fi
                         ocf_log info "Adding IPv4 address $addr to $dev"
                 fi
                 if [ "$1" = "del" ]; then
ocsi1# patch -p0 <  /tmp/ip.sh-netmask.patch
patching file ip.sh
ocsi1# clusvcadm -R ppr1
Member ocsi1-cluster trying to restart ppr1...success
ocsi1#  ip add sh|grep  -B 3 '10.10.21.138'
     link/ether 00:19:bb:3b:ac:89 brd ff:ff:ff:ff:ff:ff
     inet 10.10.21.132/26 brd 10.10.21.191 scope global bond0
     inet 10.10.21.142/32 scope global bond0
     inet 10.10.21.138/26 scope global secondary bond0

It works now and I know why it didn't work before.
The last time I saved a copy of the ip.sh under /usr/share/cluster,  
reading you comment bellow I can guess that the backup file was used  
and not the patched file, even the backup file had a name ip.sh-bak.

I am sorry but I did expect from the cluster to pick up the backup file.
I should haven't saved the backup file under the same directory.

Thanks for your help.




>
> If there are any of these, you need to chmod -x them or remove them.  In
> later versions, we will ignore some obvious ones (.rpm*, for example).
>
> -- Lon
>
> --
> Linux-cluster mailing list
> Linux-cluster at redhat.com
> https://www.redhat.com/mailman/listinfo/linux-cluster
>







More information about the Linux-cluster mailing list