IPMultiNIC fail

[knguessan]

Hi all

I try to configure an ipmultinicB ressource but it don't work properly. All my NICB are working but ont ipmultiniB.
my files configuration solaris 9 are under.

root@srvcirp1resto-01 # cat hostname.ce1
10.242.69.180 netmask 255.255.255.0 broadcast + deprecated -failover up
root@srvcirp1resto-01 #
root@srvcirp1resto-01 # cat hostname.ce3
10.242.69.181 netmask 255.255.255.0 broadcast + deprecated -failover up \
addif 10.242.69.94 netmask 255.255.255.0 broadcast + up
root@srvcirp1resto-01 #
root@srvcirp1resto-01 #
root@srvcirp1resto-02 #
root@srvcirp1resto-02 # cat hostname.ce1
10.242.69.183 netmask 255.255.255.0 broadcast + deprecated -failover up
root@srvcirp1resto-02 #
root@srvcirp1resto-02 # cat  hostname.ce3
10.242.69.182 netmask 255.255.255.0 broadcast + deprecated -failover up \
addif 10.242.69.95 netmask 255.255.255.0 broadcast + up
root@srvcirp1resto-02 #

When i check the IPmultinicB agent with haagent i have this output.

root@srvcirp1resto-02 # haagent -display IPMultiNICB
#Agent       Attribute  Value
IPMultiNICB  AgentFile 
IPMultiNICB  Faults     0
IPMultiNICB  Running    No
IPMultiNICB  Started    No
root@srvcirp1resto-02 #

I think the agent don't start.

root@srvcirp1resto-02 # ps -ef |grep -i ip
    root  3014     1  0 15:08:15 ?        0:00 /opt/VRTSvcs/bin/IP/IPAgent -type IP
    root 22669 14846  0 15:53:48 pts/3    0:00 grep -i ip
root@srvcirp1resto-02 #



Some body could help me to resolve this.

Regards

NK.

[Dev_Roy]

Checklist to ensure the proper operation of MultiNICB
For the MultiNICB agent to function properly, you must satisfy each item in the following list:
■
Each interface must have a unique MAC address.
■
A MultiNICB resource controls all the interfaces on one IP subnet.
■
At boot time, you must configure and connect all the interfaces that are under the MultiNICB resource and give them test IP addresses.
■
All test IP addresses for the MultiNICB resource must belong to the same subnet as the virtual IP address.
■
Reserve the base IP addresses, which the agent uses to test the link status, for use by the agent. These IP addresses do not get failed over.
■
The IgnoreLinkStatus attribute is set to 1 (default) when using trunked interfaces.
■
If you specify the NetworkHosts attribute, then that host must be on the same subnet as the other IP addresses for the MultiNICB resource.
■
Test IP addresses have "nofailover" and "deprecated" flags set at boot time.
■
/etc/default/mpathd/ has TRACK_INTERFACES_ONLY_WITH_GROUPS=yes.
■
If you are not using Solaris in.mpathd, all MultiNICB resources on the system have the UseMpathd attribute set to 0 (default). You cannot run in.mpathd on this system.
■
If you are using Solaris in.mpathd, all MultiNICB resources on the system have the UseMpathd attribute set to 1.

[Dev_Roy]

For more information look at the bundle agent guide:
http://sfdoccentral.symantec.com/sf/5.0MP3/solaris/pdf/vcs_bundled_agents.pdf

[knguessan]

I use Storage foundation 4.1 and i recently applied PM2 of 4.1

[Dev_Roy]

Look at the bundle agent guide of 4.1 from DocCentral http://sfdoccentral.symantec.com/sf/4.1/solaris/pdf/vcs_bundled_agents.pdf
In either case "Test IP addresses have "nofailover" and "deprecated" flags set at boot time"

[g_lee]

If you recently upgraded from 4.1 to 4.1MP2 - did you copy the new types.cf after running installmp / installing the patches as instructed in the release notes?

VERITAS Cluster Server (tm) 4.1 Maintenance Pack 2 (Solaris) - Release Notes
http://seer.entsupport.symantec.com/docs/287683.htm

relevant step on page 12:
9. Copy the new types.cf file from /etc/VRTSvcs/conf to /etc/VRTSvcs/conf/config.
# cp /etc/VRTSvcs/conf/types.cf /etc/VRTSvcs/conf/config
Note If you have customized entries in the types.cf file from the previous version, you may transfer them to the new types.cf file from the backup file created prior to patch installation.

Various agents (including IPMultiNICB) were changed between 4.1 and 4.1MP2, so if the types.cf file is not updated per instructions, it will not be able to probe as the type definition won't have the required/correct attributes that the agent binary is looking for - hence it won't start/it won't probe. (The file is not copied automatically with the patch installation as if users have customised type definitions/have set non-default values for types then these would be lost - hence a step is included in the instructions to copy manually / add the new attributes as required.)

To fix your issue now: Stop cluster, make sure the types.cf has the required updates for 4.1MP2, then start cluster again and it should be able to start/probe if the attributes have been updated correctly.

[g_lee]

fyi Dev,
The flags you've mentioned are already set  here - as seen in the hostname.ceX contents provided:

root@srvcirp1resto-01 # cat hostname.ce1
10.242.69.180 netmask 255.255.255.0 broadcast + deprecated -failover up

root@srvcirp1resto-01 # cat hostname.ce3
10.242.69.181 netmask 255.255.255.0 broadcast + deprecated -failover up \
addif 10.242.69.94 netmask 255.255.255.0 broadcast + up

see the flags "deprecated" (1st line of each file, 6th field), "-failover" (7th field) (the ifconfig option is [failover | -failover], so minus failover as opposed to "failover" = nofailover/not failover)

[Dev_Roy]

My appologies mate, I just forgot the significance of the "-" sign. I hope rest all the things are in place as outlined in bundle agent. Even I comply with your suggestion in this case, would suggest user to replace the types.cf using following procedure

# hastop -all -force
#mv /etc/VRTSvcs/conf/config/types.cf /etc/VRTSvcs/conf/config/types.cf.old
#cp  /etc/VRTSvcs/conf/types.cf  /etc/VRTSvcs/conf/config/types.cf
# hastart --> on all the nodes.

Regards,
Dev

[knguessan]

Hi all

It works now i copy types.cf on my old backup on system.

Thank's you for yor help.