Brocade Fibre Channel Networking Community

Expand all | Collapse all

48000 CP failover question

  • 1.  48000 CP failover question

    Posted 03-15-2013 10:18 AM

    We have a 48000 switch that has the local CP in standby state and the remote CP in active state due to a previous event that has been corrected. Will power cycling it revert back to normal ? IE... local CP to Active ?

    Also, Approx. how long will the switch take to come up after a power cycle ?

    Thanks.


    #BrocadeFibreChannelNetworkingCommunity


  • 2.  Re: 48000 CP failover question

    Posted 03-15-2013 10:30 AM

    CP's on 48000 are located on Slot 5 and 6.

    One of both is Active the other Standby.

    No matter if i.e. Slot 5 is Active and Slot 6 is in Standby or Viceversa, none of both Slot have any priority.

    what is the reason you want to see any Specific as Active ?


    #BrocadeFibreChannelNetworkingCommunity


  • 3.  Re: 48000 CP failover question

    Posted 03-15-2013 10:42 AM

    The "local" and "remote" designations are only a function of which CP the user logs into (which of the 3 IP addresses they specified -- there is one for each CP plus the overall chassis IP). It sounds like here the user logged in using the IP of the standby CP for some reason (rather than the chassis IP).

    A power cycle should not be necessary to switch roles (in fact it probably would not do so). The 'hafailover' command can be used to promote the current standby to the active state. But even that shouldn't really be necessary. If they just log in using the chassis IP then they should see the "local" as the active CP.

    Example:

    bro48000:admin> hashow

    Local CP (Slot 5, CP0): Active, Warm Recovered

    Remote CP (Slot 6, CP1): Standby, Healthy

    HA enabled, Heartbeat Up, HA State synchronized

    bro48000:admin> hafailover

    Local CP (Slot 5, CP0): Active, Warm Recovered

    Remote CP (Slot 6, CP1): Standby, Healthy

    HA enabled, Heartbeat Up, HA State synchronized

    Warning: This command is being run on a redundant control processor(CP) system, and this operation will cause the active CP to reset. Therefore all existing telnet sessions are required to be restarted.

    Are you sure you want to fail over to the standby CP ? y

    Forcing Failover ...

    Connection to bro48000-40 closed by foreign host.

    bro48000:admin> hashow

    Local CP (Slot 6, CP1): Active, Warm Recovered

    Remote CP (Slot 5, CP0): Non-Redundant


    #BrocadeFibreChannelNetworkingCommunity


  • 4.  Re: 48000 CP failover question

    Posted 03-15-2013 11:13 AM

    Can't ping or login to Chassis I/P or CP1 I/P. Can only ping and log into CP0

    hashow when logged into CP0 lists CP0 as local / standby and CP1 as Remote / Active

    We suspect this state happened because a network switch came up after the 48000 reboot after a power outage.


    #BrocadeFibreChannelNetworkingCommunity


  • 5.  Re: 48000 CP failover question

    Posted 03-15-2013 01:04 PM

    I suspect a reboot or power cycle will get us out of this situation. We need to request a window to do that and were wondering how long the 48000 takes to reboot and/or power cycle.

    thanks


    #BrocadeFibreChannelNetworkingCommunity


  • 6.  Re: 48000 CP failover question

    Posted 03-16-2013 06:24 AM

    Do you have both CP cards plugged into your network?   IP addresses set correctly?  are you using a gateway on you network?... 

    Need to get both CP cards on the network,, the chassis ip float back and forth between the two cards, depending  one which card is active. 

    Is you network using any kind of port security, only allowing one MAC address per port?   

    You have to get you network connectivity corrected, where you can talk to all three ip address.  If you don't do that it will also make it very hard to perform any firmware upgrades to the switch.      "GOOD LUCK"


    #BrocadeFibreChannelNetworkingCommunity


  • 7.  Re: 48000 CP failover question

    Posted 03-28-2014 12:06 AM
    Hi How to check CP failover logs...and what is the meaning of WARM Recovered and COLD Recovered in hashow command
    #BrocadeFibreChannelNetworkingCommunity


  • 8.  Re: 48000 CP failover question

    Posted 03-28-2014 12:25 AM

    you have WARM recovered i.e. when executed the command hafailover or haenable

     

    about,

     

    ->How to check CP failover logs....

     

    I must admit, I've never check such logs, and to be honestly I'm not sure if sany such features is available.

     

    I'll check on one my 48K in the lab about such failover logs.


    #BrocadeFibreChannelNetworkingCommunity


  • 9.  Re: 48000 CP failover question

    Posted 03-31-2014 01:31 PM

    Failovers generate an event in the switch event logs configured by default. Log into the gui and you'll generally see it at the bottom right :)


    #BrocadeFibreChannelNetworkingCommunity