Cluster Suite IP-Aliasing

 
Thread Tools Search this Thread
Operating Systems Linux Red Hat Cluster Suite IP-Aliasing
# 1  
Old 05-05-2010
Cluster Suite IP-Aliasing

Hi,

is it normal, that the IP alias (service IP) can't be seen with ifconfig -a , as eth0:1 for example

the IP is on the node, you can ping it, and open ports for that IP

look at this:

Code:
[root@uscltest3 ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000
    link/ether 00:50:56:95:13:2c brd ff:ff:ff:ff:ff:ff
    inet 10.200.218.82/24 brd 10.200.218.255 scope global eth0
    inet 10.200.218.85/24 scope global secondary eth0
    inet6 fe80::250:56ff:fe95:132c/64 scope link 
       valid_lft forever preferred_lft forever
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000
    link/ether 00:50:56:95:2e:f5 brd ff:ff:ff:ff:ff:ff
    inet 192.168.0.82/24 brd 192.168.0.255 scope global eth1
    inet6 fe80::250:56ff:fe95:2ef5/64 scope link 
       valid_lft forever preferred_lft forever
4: sit0: <NOARP> mtu 1480 qdisc noop 
    link/sit 0.0.0.0 brd 0.0.0.0
5: virbr0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue 
    link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
    inet 192.168.122.1/24 brd 192.168.122.255 scope global virbr0
    inet6 fe80::200:ff:fe00:0/64 scope link 
       valid_lft forever preferred_lft forever
[root@uscltest3 ~]# ifconfig -a
eth0      Link encap:Ethernet  HWaddr 00:50:56:95:13:2C  
          inet addr:10.200.218.82  Bcast:10.200.218.255  Mask:255.255.255.0
          inet6 addr: fe80::250:56ff:fe95:132c/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:40389 errors:0 dropped:0 overruns:0 frame:0
          TX packets:21471 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:12202049 (11.6 MiB)  TX bytes:9060670 (8.6 MiB)
          Base address:0x2000 Memory:d8920000-d8940000 

eth1      Link encap:Ethernet  HWaddr 00:50:56:95:2E:F5  
          inet addr:192.168.0.82  Bcast:192.168.0.255  Mask:255.255.255.0
          inet6 addr: fe80::250:56ff:fe95:2ef5/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:213 errors:0 dropped:0 overruns:0 frame:0
          TX packets:56 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:20831 (20.3 KiB)  TX bytes:10360 (10.1 KiB)
          Base address:0x2040 Memory:d8940000-d8960000 

lo        Link encap:Local Loopback  
          inet addr:127.0.0.1  Mask:255.0.0.0
          inet6 addr: ::1/128 Scope:Host
          UP LOOPBACK RUNNING  MTU:16436  Metric:1
          RX packets:5927 errors:0 dropped:0 overruns:0 frame:0
          TX packets:5927 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0 
          RX bytes:7709874 (7.3 MiB)  TX bytes:7709874 (7.3 MiB)

sit0      Link encap:IPv6-in-IPv4  
          NOARP  MTU:1480  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0 
          RX bytes:0 (0.0 b)  TX bytes:0 (0.0 b)

virbr0    Link encap:Ethernet  HWaddr 00:00:00:00:00:00  
          inet addr:192.168.122.1  Bcast:192.168.122.255  Mask:255.255.255.0
          inet6 addr: fe80::200:ff:fe00:0/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:54 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0 
          RX bytes:0 (0.0 b)  TX bytes:10680 (10.4 KiB)

# 2  
Old 05-05-2010
Not really. How did you set up your virtual interface?
# 3  
Old 05-06-2010
with the lucy http interface

defined a resource and put it to the service, I think there were ip and netmask to configure, not more


I'll post my cluster.conf tomorrow



do you have a red hat cluster with a service ip, where you can see the ip with ifconfig?

---------- Post updated 06-05-10 at 09:55 ---------- Previous update was 05-05-10 at 19:33 ----------

I worked with AIX hacmp and SUN Cluster previously, and I must say the Red Hat Cluster Suite seems to be very buggy


here is the cluster.conf

Code:
[root@uscltest2 ~]# cat /etc/cluster/cluster.conf
<?xml version="1.0"?>
<cluster alias="uscltest" config_version="38" name="uscltest">
        <fence_daemon clean_start="0" post_fail_delay="0" post_join_delay="3"/>
        <clusternodes>
                <clusternode name="uscltest3" nodeid="1" votes="1">
                        <fence>
                                <method name="1">
                                        <device name="scsi-fench" node="uscltest3"/>
                                </method>
                        </fence>
                </clusternode>
                <clusternode name="uscltest2" nodeid="2" votes="1">
                        <fence>
                                <method name="1">
                                        <device name="scsi-fench" node="uscltest2"/>
                                </method>
                        </fence>
                </clusternode>
                <clusternode name="uscltest1" nodeid="3" votes="1">
                        <fence>
                                <method name="1">
                                        <device name="scsi-fench" node="uscltest1"/>
                                </method>
                        </fence>
                </clusternode>
        </clusternodes>
        <cman/>
        <fencedevices>
                <fencedevice agent="fence_scsi" name="scsi-fench"/>
        </fencedevices>
        <rm>
                <failoverdomains>
                        <failoverdomain name="domain3" nofailback="0" ordered="1" restricted="1">
                                <failoverdomainnode name="uscltest3" priority="1"/>
                                <failoverdomainnode name="uscltest2" priority="2"/>
                                <failoverdomainnode name="uscltest1" priority="3"/>
                        </failoverdomain>
                        <failoverdomain name="domain2" ordered="1" restricted="1">
                                <failoverdomainnode name="uscltest1" priority="1"/>
                                <failoverdomainnode name="uscltest2" priority="1"/>
                                <failoverdomainnode name="uscltest3" priority="2"/>
                        </failoverdomain>
                        <failoverdomain name="domain1" ordered="1" restricted="1">
                                <failoverdomainnode name="uscltest1" priority="1"/>
                                <failoverdomainnode name="uscltest3" priority="3"/>
                                <failoverdomainnode name="uscltest2" priority="2"/>
                        </failoverdomain>
                </failoverdomains>
                <resources>
                        <ip address="10.200.218.85" monitor_link="1"/>
                        <lvm lv_name="test3lv" name="test3lvm" vg_name="test3vg"/>
                        <fs device="/dev/test3vg/test3lv" force_fsck="0" force_unmount="1" fsid="921" fstype="ext3" mountpoint="/cluster/test3" name="test3fs" self_fence="0"/>
                        <script file="/cluster/test3/httpd/bin/apachectl" name="test3_http"/>
                </resources>
                <service autostart="1" domain="domain3" exclusive="0" name="test3" recovery="relocate">
                        <ip ref="10.200.218.85">
                                <lvm ref="test3lvm">
                                        <fs ref="test3fs">
                                                <script ref="test3_http"/>
                                        </fs>
                                </lvm>
                        </ip>
                </service>
        </rm>
</cluster>

# 4  
Old 06-15-2010
Short answer: this is perfectly normal. What happened was that an IP address was added to the interface without a differentiating label.

Believe it or not, the "ifconfig" application is strongly discouraged because it has a tendency to dramatize situations like this. Smilie It would probably have been removed years ago if not for legacy support.

Anyway, if you want to understand a bit more about what is happening, take a look at this iproute2 documentation. They explain that although all alias addresses are secondary addresses, the inverse is not necessarily true.

Then try the following:
Code:
[root@uscltest2 ~]# ip addr show dev eth0
[root@uscltest2 ~]# ip addr add 10.200.218.86/24 broadcast 10.200.218.255 dev eth0
[root@uscltest2 ~]# ip addr
[root@uscltest2 ~]# ifconfig
[root@uscltest2 ~]# ip addr add 10.200.218.87/24 broadcast 10.200.218.255 label eth0:alias dev eth0
[root@uscltest2 ~]# ip addr
[root@uscltest2 ~]# ifconfig

You will notice that after the first add, the output of "ip addr" showed the address, but the output of "ifconfig" did not. Additionally, when you added a label the second time, the output of "ip addr" still shows the address, but the output of "ifconfig" also show our new label. (which only says "eth0:alias" because we told it to - it could literally be anything)

After this little demo, reset things to normal:
Code:
[root@uscltest2 ~]# ip addr del 10.200.218.86/24 dev eth0
[root@uscltest2 ~]# ip addr del 10.200.218.87/24 dev eth0

This User Gave Thanks to jjinno For This Post:
# 5  
Old 06-22-2010
you are right jjinno, thank you for this hint

I wonder why the don't add a label when applying the alias from the cluster suite, I guess I'm not the only one who is confused about that Smilie
Login or Register to Ask a Question

Previous Thread | Next Thread

9 More Discussions You Might Find Interesting

1. Solaris

Directory Aliasing

Hello, I need to rediret the directory location at run time as:- If I type cd /dir1/dir2 then it should take me to /dir1/dir3 at run time. Inside dir1, both dir2 and dir3 directories are present. Could anyone guild me over it? Thanks Shubham (5 Replies)
Discussion started by: shubh05
5 Replies

2. Red Hat

How can i configure fence_cisco_ucs with redhat cluster suite

HI How can i configure fence_cisco_ucs with redhat cluster suite. i was trying to configure fencing aganet with red ahat cluster suite i issued command # /usr/sbin/fence_vmware_soap -a 172.22.90.61 -l admin -p duc2Cisco -o reboot then i got below error: Failed: You have to... (0 Replies)
Discussion started by: bentech4u
0 Replies

3. Homework & Coursework Questions

aliasing command

hi to all members in this forums .. Nice to meet you... i only have a questions about command utility of "dd": for example i will type: pico trial this is my file. and save it and type chmod a+x trial can there be a way that instead of typing: dd if=trial of=trial.copy conv=ucase... (3 Replies)
Discussion started by: marahtia
3 Replies

4. UNIX for Dummies Questions & Answers

How can I do aliasing in shellscript?

#Example.sh alias rmv 'sh Example2.sh' when i execute exapme.sh alias name not working. how i solve this problem?? (9 Replies)
Discussion started by: arun508.gatike
9 Replies

5. High Performance Computing

Solaris 10 - Cluster Suite 3.2 2/08 prob

Hi, I am relatively new to Sun Solaris. I am installing 2 node cluster and when I was installing cluster suite 3.2 2/08, it is giving me this error as "Conflict - Incomplete version detected". Please help me in this regard and also suggest me any valuable point to consider before going ahead... (3 Replies)
Discussion started by: hi2joshi
3 Replies

6. UNIX for Advanced & Expert Users

aliasing the TO name in Unix mails

HI All, I have a small question/clarification/doubt . Does anyone know how to provide alias in the mails that come from Unix servers. Like i have a Unix server that generates status mails every hour for a scheduled job , it sends me mails with status with the ID in from field as... (3 Replies)
Discussion started by: bsandeep_80
3 Replies

7. UNIX for Dummies Questions & Answers

What is IP aliasing ?

Hi, Can anyone tell me what is IP aliasing ? I do not know if this is the right place for this thread !!. Thanks in advance, Gideon. (2 Replies)
Discussion started by: preethgideon
2 Replies

8. UNIX for Dummies Questions & Answers

Basic regarding aliasing

Hi All, One small request. In UNIX/LINUX we can have our own aliases in .bashrc file. My doubt is when we add a new alias and if it is tried in already opened terminals it will not work, may be it is not going to recognise. Bit when we use the same alias in new terminal it will work. I... (1 Reply)
Discussion started by: chanakyahere
1 Replies

9. Linux

RedHat Cluster Suite - multicast problem

Hi, I have a Redhat Cluster Suite with 2 nodes, using option "Multicast HeartBeating" on clumembd, whith default IP: 225.0.0.11 The cluster member names are: # grep node /etc/hosts 192.168.100.1 node1 192.168.100.2 node2 My problem is that the Multicast IP is relating to other network... (1 Reply)
Discussion started by: Livio
1 Replies
Login or Register to Ask a Question