12-01-2011 02:18 AM
Need your help.
We have a WX deployment that's working fine for our client. However, we'v noticed that the WX (Hub at the Head quarters) has restarted the core switch on two different occassions: 1) First was when we restarted the WX after making some configuration changes. 2) When the cable connecting the switch to the WX (local interface only) was disconnected (customer removed this cable for troubleshooting purposes); Note that the cable was removed from the WX local interface and NOT at the core switch.
This is a bank with over 200 sites and the customers are very very skeptical because the core switch connects these sites and they can't afford any downtime.
PS: WX is currently connected to 10 of these sites in a hub & spoke topology. Off-path mode with WCCP inbound re-direction was applied with WXOS 18.104.22.168 used across all sites.
What could be the cause of this strange behaviour?
Thanks in advance for your swift response.
12-01-2011 07:41 PM
When you powering off or unplug WXC from network it cannot trigger/initiate any signal to external device to reload.
When you remove WXC from network, the WCCP neighboring with WXC will be broken on core switch, So I suspect WCCP protocol might trigger anonymous reboot of core switch.
If core switch reboots due to an know/unknown IOS caveat/bug or software failure, then it does create crachinfo before it reloads.
Check logging errors; if core switch reloads then it would be erased from switch, so if customer has syslog server configured, then check whether any abnormal logs appearing before core switch reload.
It will help you to narrow down the issue.
If crashinfo was created on core switch, then I suggest you to analysis the issue switch TAC team.