Support Forum
The Forums are a place to find answers on a range of Fortinet products from peers and product experts.
dilic
New Contributor II

Promote Slave to Master within 2 node failover HA cluster?

Hello.

Is there an easy way to promote or change HA roles from SLAVE to MASTER in Fortigate (800c) 2 nodes HA failover cluster, within Fortigate gui or cmd commands/configuration?  

2 Solutions
Robin_Svanberg
Contributor

dilic wrote:

Hello.

Is there an easy way to promote or change HA roles from SLAVE to MASTER in Fortigate (800c) 2 nodes HA failover cluster, within Fortigate gui or cmd commands/configuration?  

Per default (If you haven´t enabled device priority override") the HA Master election is based on the following:

 

1. Monitored port

2. System Uptime, the one that has been up for the longest time

3. Unit Priority, the one with the highest priority is master

4. Serialnumber, the highest serial number is master

 

The easiest way is to reset the uptime on the master by running the command "diagnose sys ha reset-uptime". If you change the priority a failover will not occur.

 

If you have enabled device priority override the system uptime isn´t part of the election and it will in that case use the Unit priority number.

 

A reboot of the primary unit will also cause a failover :) But if you have enabled device priority override the unit with the highest priority will be the master when it´s back online.

 

 

Robin Svanberg Network Consultant @ Ethersec AB in Östersund, Sweden

 

robin.svanberg@ethersec.se

View solution in original post

Robin Svanberg Network Consultant @ Ethersec AB in Östersund, Sweden robin.svanberg@ethersec.se
dilic

Robin Svanberg wrote:

dilic wrote:

Hello.

Is there an easy way to promote or change HA roles from SLAVE to MASTER in Fortigate (800c) 2 nodes HA failover cluster, within Fortigate gui or cmd commands/configuration?  

Per default (If you haven´t enabled device priority override") the HA Master election is based on the following:

 

1. Monitored port

2. System Uptime, the one that has been up for the longest time

3. Unit Priority, the one with the highest priority is master

4. Serialnumber, the highest serial number is master

 

The easiest way is to reset the uptime on the master by running the command "diagnose sys ha reset-uptime". If you change the priority a failover will not occur.

 

If you have enabled device priority override the system uptime isn´t part of the election and it will in that case use the Unit priority number.

 

A reboot of the primary unit will also cause a failover :) But if you have enabled device priority override the unit with the highest priority will be the master when it´s back online.

 

Hmm,

I didn't enabled device priority override. So, if  I just execute "diagnose sys ha reset-uptime" in cmd shell, that will do the trick. Cause I'm on the MASTER node by default in HA failover cluster.  

 

Or I'm must go first on Master node by executing commands:

 

#config global

# get system ha status

     Model: FortiGate-800C      Mode: a-p      Group: 0      Debug: 0      ses_pickup: enable, ses_pickup_delay=disable      Master:130 CWa01 FG800Cxxxxxxxxx7 1      Slave :140 CWb01 FG800Cxxxxxxxxx9 0      number of vcluster: 1      vcluster 1: work 169.254.0.2      Master:0 FG800Cxxxxxxxxx7      Slave :1 FG800Cxxxxxxxxx9

# exec ha manage 1

# diagnose sys ha reset-uptime

 

Please confirm right procedure ...

As you can see, node which I want to promote to MASTER, has alredy  higher priority (140) ...

 

 

 

 

View solution in original post

12 REPLIES 12
iJake
Contributor

I'm not sure if there's an easy way to do this, but you can set the priority higher on your secondary and enable device priority override?

 

The document below has details about HA override

http://docs-legacy.fortin...lp/HA_FGCP.081.30.html

......

-Jake

...... -Jake
emnoc
Esteemed Contributor III

Yes adjust the  HA priority so the now slave is higher than the now-active

 

I have "    set priority 255  " on my active and "    set priority 100 " on my  slave, so if I set the  priority to 99, the slave would be active if all of the monitor interface are in an up and ready

 

 

PCNSE 

NSE 

StrongSwan  

PCNSE NSE StrongSwan
Robin_Svanberg
Contributor

dilic wrote:

Hello.

Is there an easy way to promote or change HA roles from SLAVE to MASTER in Fortigate (800c) 2 nodes HA failover cluster, within Fortigate gui or cmd commands/configuration?  

Per default (If you haven´t enabled device priority override") the HA Master election is based on the following:

 

1. Monitored port

2. System Uptime, the one that has been up for the longest time

3. Unit Priority, the one with the highest priority is master

4. Serialnumber, the highest serial number is master

 

The easiest way is to reset the uptime on the master by running the command "diagnose sys ha reset-uptime". If you change the priority a failover will not occur.

 

If you have enabled device priority override the system uptime isn´t part of the election and it will in that case use the Unit priority number.

 

A reboot of the primary unit will also cause a failover :) But if you have enabled device priority override the unit with the highest priority will be the master when it´s back online.

 

 

Robin Svanberg Network Consultant @ Ethersec AB in Östersund, Sweden

 

robin.svanberg@ethersec.se

Robin Svanberg Network Consultant @ Ethersec AB in Östersund, Sweden robin.svanberg@ethersec.se
dilic

Robin Svanberg wrote:

dilic wrote:

Hello.

Is there an easy way to promote or change HA roles from SLAVE to MASTER in Fortigate (800c) 2 nodes HA failover cluster, within Fortigate gui or cmd commands/configuration?  

Per default (If you haven´t enabled device priority override") the HA Master election is based on the following:

 

1. Monitored port

2. System Uptime, the one that has been up for the longest time

3. Unit Priority, the one with the highest priority is master

4. Serialnumber, the highest serial number is master

 

The easiest way is to reset the uptime on the master by running the command "diagnose sys ha reset-uptime". If you change the priority a failover will not occur.

 

If you have enabled device priority override the system uptime isn´t part of the election and it will in that case use the Unit priority number.

 

A reboot of the primary unit will also cause a failover :) But if you have enabled device priority override the unit with the highest priority will be the master when it´s back online.

 

Hmm,

I didn't enabled device priority override. So, if  I just execute "diagnose sys ha reset-uptime" in cmd shell, that will do the trick. Cause I'm on the MASTER node by default in HA failover cluster.  

 

Or I'm must go first on Master node by executing commands:

 

#config global

# get system ha status

     Model: FortiGate-800C      Mode: a-p      Group: 0      Debug: 0      ses_pickup: enable, ses_pickup_delay=disable      Master:130 CWa01 FG800Cxxxxxxxxx7 1      Slave :140 CWb01 FG800Cxxxxxxxxx9 0      number of vcluster: 1      vcluster 1: work 169.254.0.2      Master:0 FG800Cxxxxxxxxx7      Slave :1 FG800Cxxxxxxxxx9

# exec ha manage 1

# diagnose sys ha reset-uptime

 

Please confirm right procedure ...

As you can see, node which I want to promote to MASTER, has alredy  higher priority (140) ...

 

 

 

 

emnoc
Esteemed Contributor III

Do it on the current master, that's all that you have to do and the new master will be selected.

 

ken

PCNSE 

NSE 

StrongSwan  

PCNSE NSE StrongSwan
dilic
New Contributor II

Thank you ...

Command "diagnose sys ha reset-uptime" did the trick ...

howardsinc

I know this is solved but here is a handy trick for the lab.

 

while you are testing only, a useful command to force a unit to become the Master is:

 

'diagnose sys ha set-as-master enable'

 

(page 43) this command is only to be used in a lab environment. 

http://docs.fortinet.com/...igate-ha-526.pdf 

JNCIA, CCNP R/S, NSE4 , NSE7, Associate of (ISC)²

JNCIA, CCNP R/S, NSE4 , NSE7, Associate of (ISC)²
ede_pfau
Esteemed Contributor III

'set-as-master' is in v5.2.5 and newer only.

 

Does it actually trigger a failover, or does it only 'enable' the slave unit to become master when failing over the next time?


Ede

"Kernel panic: Aiee, killing interrupt handler!"
Ede"Kernel panic: Aiee, killing interrupt handler!"
netmin

It enforces a failover...I had posted an example some time ago, here: https://forum.fortinet.com/FindPost/113598

 

The 5.2.0 What's New Guide mentions it in short:

"HA failover can now be enabled and disabled using the following CLI commands: • diagnose sys ha set-as-master enable: immediately enables the local FortiGate unit as the HA master."

Labels
Top Kudoed Authors