@AhmedKamal, the WAN IP address of master and slave devices will be same. If you have chance to look on the slave device, all it’s WANs will be in Standby mode without IP address. Once the failover occurs, WANs of slave device will be transition to active and using the same IP address as master device.
Thanks for the helpful answer.
So what about VPN profiles especially those speedfusion profiles created on remote site boxes and points at “remote ID” of one of the HQ boxes?
After adding the second box in HQ and configuring HA, do I have to configure a new profile on the remote site boxes with the new box ID?
No. Slave unit always sync the config from the master automatically.
Does HA support in peplink balance if running on Layer 2 SF VPN environment?
Yes, it’s supported.
Dear Sit Loong, thanks for the prompt confirmation, because we didn’t try before if peplink running L2SF VPN but add in one more peplink as HA. I believe the configuration wise should be the same as standard right?
Yes, it’s the same.
I recently purchased 2 units of Balance 1350, updated Firmware to version 7.1.0 build 2287. I configured the first unit with Drop in mode. But I cannot enable HA as it says ‘High Availability is disabled because the LAN bypass port is in drop-in mode’.
I want to configure HA and use drop in mode.
Make sure you define the drop in interface not using the LAN by pass interface then you can enable the HA settings for the devices.
apart from lost connectivity in all of the WAN interfaces to “outside”, what else are the reason to have a failover on the system, to change from master to slave?.
I tested this on LAB and i only found that removing or the LAN (was using only one cable) or removing WANs cable (i was using 2 ISP and i removed both, only 1 doesnt make a failover) i generate a failover on the system.
There some other way to test the failover?.
Disconnect the power chord. No joke:wink:
Any plans/possibility to have this on the Balance 30 LTE?
I have a similar situation here and would like to ask the same question to peplink, is it required to have 2 switches attached to the ISP router? What will be the difference with and without the switches?
I would use the Switches.
- to be shure, that the ARP-table is correct
- to be shure, that the Port config is correct
(If ther is eg. a routing entry in the ISP-router like “send all data on LAN1”, then you will have a problem)
- I certainly forgot some points
The Switches are not 100% necessary, but it easier with them. If you don’t want to use them, you have to understand the configuration and the routing mechanism of your ISP router
Thanks for your quick reply, some points like for example arp or vlan database would serve the purpose of those switches over there. Thanks for highlighting this to me.
Can anyone please confirm if these are the scenarios in which the slave unit will become master? Am I missing one?
- In case of total failure
- LAN port disconnection
- All enabled wan ports disconnection
TK, please correct me if I’m wrong:
The slave will become master if the master stops answering on the virtual ip.
The master will stop answering on the virtual ip if :
- In case of total failure
- In case of all WAN disconnected
The slave will become a master even if the master doesn’t become a slave if :
- there are LAN port disconnection (he doesn’t see the virtual ip anymore)
- the switch has issues etc
We’re encountering deployment issues because other brands can have virtual ip’s per vlan and selective master/slave scenarios per subnet/vlan. We’re hoping Peplink will develop it further. This is a critical feature for Entreprise SDWAN.