We need to configure Ignition Redundancy for one of our client in couple days. We gone through Ignition Redundancy Setup video on Inductive University and learned little about it.
We tested the Redundancy on our local system (before actual configuration on Client’s site) and wanted to confirm our Configuration. Following are our configuration details -
[ul]1. We installed Ignition on two systems (Ex: 192.168.0.101, 192.168.0.102).
2. 192.168.0.101 Ignition System defined as Mode = Master
Standby Activity Level = Warm
Auto-detect network interface = Off
Network Bind Interface = 192.168.0.101
Autodetect HTTP Address = On
Master Node Address = 192.168.0.1013. 192.168.0.102 Ignition System defined as Mode = Backup
Standby Activity Level = Warm
Auto-detect network interface = Off
Network Bind Interface = 192.168.0.102
Autodetect HTTP Address = On
Master Node Address = 192.168.0.1014. We are keeping the Standby Activity Level as Warm as their process is critical. Remaining settings are defaults.[/ul]
We tested accessing the Project from 192.168.0.103 system and make Network Off for 192.168.0.101. It takes few seconds of time (it must be Failover Timeout setting) to show the Master system unavailability and switching to Backup System. When Master System is up again, the switching done immediately.
Is our way of configuration is correct in general? Our systems are connected to a bigger Company network.
Looks like you are on the right path. I would turn autodetect http off and manually enter the two ip addresses and ports for each of the servers in the http addresses section also.
[quote]Looks like you are on the right path. I would turn autodetect http off and manually enter the two ip addresses and ports for each of the servers in the http addresses section also.
[/quote]
I set up a redundant pair and did this too but when running a project or designer, I looked at the diagnostics under the help tab and found an illegal setup error complaining about this in the log viewer tab. This is the only place I saw this error. Called tech support and was told to put only the address of the individual server in address section, not both. When I did this the error went away.
I just looked at one of our redundant systems and didnt see this error.
from the manual:
even on our non redundant configurations that are out on the public I put the domain name and also the public ip address in the address list, that way if there are any dns issues, when the client is launched it will try both the domain address and if that doesnt work it will use the public ip.
I cant speak for 7.7, but this is the way I have always used these configuration settings and they have worked with no issue.