SRX

Ive configured a new reth int on a pair of clustered srx650 running junos 10.4R10.7. 3 other reth are working fine. This new one is showing as link down even though one of the two member interfaces is up.  Ive confirmed that my reth-count is set appropriately. Any ideas why an reth would show as down when one member is up?

Hello jmcgrady

What is the reth count configured?

Can you paste interface terse output?

Regards,

Raveen

Can you paste the cluster status output as well?

the reth count is 5. the reth in question is reth3.

show int terse:

ge-2/0/14               up    up
ge-2/0/14.0             up    up   aenet    --> reth3.0

ge-11/0/14              up    down
ge-11/0/14.0            up    down aenet    --> reth3.0

reth1                   up    up
reth1.0                 up    up   inet     10.1.1.1/24
reth2                   up    up
reth2.0                 up    up   inet     116.1.1.1/28
reth3                   up    down
reth3.0                 up    down inet     10.3.3.3/24

ge-11/0/14 is unplugged.

show chassis cluster status
Cluster ID: 1
Node                  Priority          Status    Preempt  Manual failover

Redundancy group: 1 , Failover count: 1
    node0                   100         primary        yes      no
    node1                   50          secondary      yes      no

Could you deactivate and activate this reth3 interface and check if it works?

deactivate interfaces reth3

commit

activate interfaces reth3

commit

Regards,

Raveen

The deactivate wasnt accepted as it complained reth3.0 didnt exist!

When i do a show int ge-2/0/14 terse on the primary node - node 0 - it replies

Interface               Admin Link Proto    Local                 Remote
ge-2/0/14               up    up
ge-2/0/14.0             up    up   aenet    --> reth3.0

But the same command when i cli onto node 1 is:

Interface               Admin Link Proto    Local                 Remote
ge-2/0/14               up    up
ge-2/0/14.0             up    up   aenet

So it looks like a mismatch between cluster members.  When i look at the config as node 1 displays it, it appears correct.

Hello

In an old post, interface-monitoring configuration has helped.

Can you try it?

http://forums.juniper.net/t5/SRX-Services-Gateway/reth-showing-down-when-only-1-member-is-shutdown/td-p/126947

Regards,

Raveen

Synchrozination of configuration is also not proper.

Please do a commit full of configuration.

Regards,

Raveen

The deactivate then activate wouldnt work because the device complained. However deleting the int definition completely and then recreating it worked.