Forum Discussion
Shyam_111786
Nimbostratus
Nimbostratusnetwork failover implementation
Hello,
I am pretty new to f5 products. I am trying to configure network failover on the bigip 3900 version 10.1. on the network failover its asking me the peer management ip and also the local and remote ip for unicast addresses. so i configure the peer management ip as the one i configure on the management port. The local and remote ip addresses are given to be the failover ip addresses.
i have configured mirrored address on the connection mirroring section as the self ip address and not the management ip address. so in the config sync section I have asked it to use the connection mirroring ip address.
apart from that i have configured a trunk and configured 1.1 and 1.2 ports as the members and made them on the external vlan. i have configured self Ip address for the external vlan.
Is there anything else i should be doing to get the network failover to work.
Thanks
hooleylistCirrostratus
Hi Shyam,
That sounds about right. Have you seen any issues when testing? I generally set up a directly connected cable between the two units to use for sync traffic, in addition to the management ports.
Aaron
Shyam_111786Hello Aaron,
Yes i have linked a cable from interface 1.8 on the active BIGIP to interface 1.8 on the standby BIGIP. I am getting errors relating to incorrect configurations when i am synching the two BIGIP's. I do have a question on this setup, when i configure the interface 1.8 dedicated for HA, it does not ask me for any IP Address. does the failover ip that i provide on the network failover section get mapped to the 1.8 interface? I have created a VLAN for 1.8 interface and named it as the HA_Heartbeat VLAN, but it still just a name. can we have a self IP Address just for the heartbeat Vlan with the same ip addresses as the failover ip addresses configured on the network failover section?
The trunk interfaces 1.1 and 1.2 (external vlan) of the active and standby BIGIPs are connected to a stacked switch (master and backup respectively)
Could you tell me what IP Address should i be giving in the config sync section? Is it the self ip address of the external vlan?
I have not yet configured an internal vlan and no floating ip addresses have been created. The internal vlan will be created at a later stage once the customer has its hosting infrastructure in place. Does this have any impact on the config synching of the bigips atleast for test purposes?
Thanks
David_McKeand_2Hi there,
99% there, you need to assign a self-ip for the HA_Heartbeat VLAN, I normally use 1.1.1.1 for the active and 1.1.1.2 for the standby devices. I would also make sure the HA_Heatbeat VLAN is untagged. Once you do this then go to the configsync setup and set the relevant ip addresses. Once you do this you will be able to sync on the dedicated cable.
I would also setup redundancy to use the HA_Heartbeat addresses as well as the mgmt addresses. Just gives you more redundancy.
Once this is all done and setup take a qk view and log into ihealth.f5.com and upload the file. This site is great and will analyse your configuration and give you tips/fixes etc.
Hope this helps, let us know how you get on.
David.- David_McKeand_2Hi there,
99% there, you need to assign a self-ip for the HA_Heartbeat VLAN, I normally use 1.1.1.1 for the active and 1.1.1.2 for the standby devices. I would also make sure the HA_Heatbeat VLAN is untagged. Once you do this then go to the configsync setup and set the relevant ip addresses. Once you do this you will be able to sync on the dedicated cable.
I would also setup redundancy to use the HA_Heartbeat addresses as well as the mgmt addresses. Just gives you more redundancy.
Once this is all done and setup take a qk view and log into ihealth.f5.com and upload the file. This site is great and will analyse your configuration and give you tips/fixes etc.
Hope this helps, let us know how you get on.
David.