Having completed a manual failover of the redundancy groups in my previous post, this test go through the process of what would have happen if we had a link fault.
Test B (Interface Failure)
In my first post creating srx cluster, I had configured Interface Monitoring. Interface monitoring can be used to trigger a failover in the event link status on an interface goes down. For this test I will be disconnecting interface ge-0/0/1, once this has been disconnected we should see that redundancy group 1 failover to Node1 from Node0.
We will check the status of the cluster and the interfaces before proceed:
{primary:node0}[email protected]_SRX220_Top# run show chassis cluster status
Cluster ID: 1
Node Priority Status Preempt Manual failover
Redundancy group: 0 , Failover count: 7
node0 100 primary no no
node1 1 secondary no no
Redundancy group: 1 , Failover count: 33
node0 100 primary yes no
node1 1 secondary yes no
{primary:node0}
[email protected]_SRX220_Top> show chassis cluster interfaces
Control link status: Up
Control interfaces:
Index Interface Status
0 fxp1 Up
Fabric link status: Up
Fabric interfaces:
Name Child-interface Status
(Physical/Monitored)
fab0 ge-0/0/5 Up / Up
fab0
fab1 ge-3/0/5 Up / Up
fab1
Redundant-ethernet Information:
Name Status Redundancy-group
reth0 Up 1
reth1 Up 1
Redundant-pseudo-interface Information:
Name Status Redundancy-group
lo0 Up 0
Interface Monitoring:
Interface Weight Status Redundancy-group
ge-0/0/2 255 Up 1
ge-3/0/2 255 Up 1
ge-3/0/1 255 Up 1
ge-0/0/1 255 Up 1
{primary:node0}
As everything is as expected, I have disconnected interface ge-0/0/1 and by running the same commands we are able to see that the link failure has been detected by running show chassis cluster interface command
{primary:node0}[edit]
[email protected]_SRX220_Top# run show chassis cluster interfaces
Control link status: Up
Control interfaces:
Index Interface Status
0 fxp1 Up
Fabric link status: Up
Fabric interfaces:
Name Child-interface Status
(Physical/Monitored)
fab0 ge-0/0/5 Up / Up
fab0
fab1 ge-3/0/5 Up / Up
fab1
Redundant-ethernet Information:
Name Status Redundancy-group
reth0 Up 1
reth1 Up 1
Redundant-pseudo-interface Information:
Name Status Redundancy-group
lo0 Up 0
Interface Monitoring:
Interface Weight Status Redundancy-group
ge-0/0/2 255 Up 1
ge-3/0/2 255 Up 1
ge-3/0/1 255 Up 1
ge-0/0/1 255 Down 1
And running show chassis cluster status we can see that redundancy group 1 has failed over from Node0 to Node1.
{primary:node0}[edit]
[email protected]_SRX220_Top# run show chassis cluster status
Cluster ID: 1
Node Priority Status Preempt Manual failover
Redundancy group: 0 , Failover count: 7
node0 100 primary no no
node1 1 secondary no no
Redundancy group: 1 , Failover count: 34
node0 0 secondary yes no
node1 1 primary yes no
As I have configured preempt on redundancy group 1, once the link (ge-0/0/1) is reconnected, it will automatically fail back to Node0
{primary:node0}[edit]
[email protected]_SRX220_Top# run show chassis cluster interfaces
Control link status: Up
Control interfaces:
Index Interface Status
0 fxp1 Up
Fabric link status: Up
Fabric interfaces:
Name Child-interface Status
(Physical/Monitored)
fab0 ge-0/0/5 Up / Up
fab0
fab1 ge-3/0/5 Up / Up
fab1
Redundant-ethernet Information:
Name Status Redundancy-group
reth0 Up 1
reth1 Up 1
Redundant-pseudo-interface Information:
Name Status Redundancy-group
lo0 Up 0
Interface Monitoring:
Interface Weight Status Redundancy-group
ge-0/0/2 255 Up 1
ge-3/0/2 255 Up 1
ge-3/0/1 255 Up 1
ge-0/0/1 255 Up 1
{primary:node0}[edit]
[email protected]_SRX220_Top# run show chassis cluster status
Cluster ID: 1
Node Priority Status Preempt Manual failover
Redundancy group: 0 , Failover count: 7
node0 100 primary no no
node1 1 secondary no no
Redundancy group: 1 , Failover count: 35
node0 100 primary yes no
node1 1 secondary yes no
My next post in this series, will demonstrate the methods of upgrading Junos version on a SRX cluster.
Keeran Marquis
Latest posts by Keeran Marquis (see all)
- Life and Times of an Unemployed Professional Speed Dater #3 - August 5, 2018
- Life and Times of an Unemployed Professional Speed Dater #2 - August 5, 2018
- Life and Times of an Unemployed Professional Speed Dater #1 - August 5, 2018
Hi,
Thanks a lot for the post.
Where do you do this testing ? on physical devices or GNS 3?
Regards,
Amar
Hi Amar
I use physical devices as I’ve got access via work 🙂
For juniper stuff if you access to an esxi server you can upload vSRX & vMXs or you can use junosphere or Cisco virl. With Cisco virl you can use junos vmdk images.
Cheers
Keeran
Hi Keeran,
I’m doing the same configuration, it works fine when one and two links are down, but when there are three links down (only one link up) I have detected a undesirable side effect with the following situation:
initially all interfaces are up
then
ge-3/0/1 goes down, no failover, primary is node0, expected
then
ge-0/0/1 goes down, no failover, primary is node0, expected
then
ge-0/0/2 goes down, no failover, primary is node0, NOT expected
My intention is when there are no interfaces available in node0 should failover to node1, even the reth have only one link up (ge-3/0/2), but in my tests this is not happening.
Could you test it? Any ideas?
Hi, finally I solved the issue with this configuration, I reproduce here just in case is useful for someone.
The idea is to trigger an event when reth0 is down, the action is to change the weights of the interfaces, then reth0 is up again even if there is only one physical interface up.
event-options {
policy keep_one_interface_reth0 {
events snmp_trap_link_down;
within 1 events snmp_trap_link_down;
attributes-match {
snmp_trap_link_down.interface-name matches reth0;
}
then {
change-configuration {
commands {
"set chassis cluster redundancy-group 1 interface-monitor ge-0/0/1 weight 128";
"set chassis cluster redundancy-group 1 interface-monitor ge-0/0/2 weight 128";
"set chassis cluster redundancy-group 1 interface-monitor ge-3/0/1 weight 128";
"set chassis cluster redundancy-group 1 interface-monitor ge-3/0/2 weight 128";
}
commit-options {
log "commited by keep_one_interface_reth0 event policy";
}
}
}
}
}
Note: in my configuration all (four) physical interfaces are assigned to reth0.