i am on ignition version 7.9.4 and have created a UDT and 200 instances based on the UDT.
In the gateway, i have created a device with blank IP address and enabled it.
An instance of the UDT will wake up at 8 am every morning and will set the ip address using t system.device.setDeviceHostname command into the device and try to connect to that device.
Like this, there are 9 devices to which , at 8 am every morning , the instances will simultaneously make the IP address change on the gateway level.
the IP addresses on the 9 devices will dynamically change every 2 minutes and it will try to connect to a new device every 2 minutes.
when i tested this, ignition crashed.
my question is , does ignition allow the user to make 9 configuration changes simultaneously at the gateway level to the devices it is connecting. or should i be doing this 1 at a time.
any help is appreciated.
What device, what driver?
FYI I stress tested in a sand box before an upgrade and created multiple (can’t remember now, but maybe up to 100?) connections to a single device, and then scripted over 20K tags to said device.
It ran for a week with 50 clients open without crashing.
8 Core CPU, 32GB RAM.
I have 9 devices with blank ip address
My instances wake up all at the same time and write an IP address into each of these devices.
It’s a Modbus TCP connection and i am changing the IP address dynamically for 9 devices at a time and trying to connect to them all at the same time.
Well I don’t have more than 1 device powered up on my test rig to run tests for you, why don’t you try a delay between each device IP write?
Note depending on the hardware, some ModBusTCP devices can only have 1 connection established to them so again… what device?
Can you upload the gateway logs that show the errors/crash happening?
Please explain why you need to do this, too, please. Why do you need to re-use device instances instead of just creating all that you need, and simply enabling them when appropriate?
i cannot upload logs unfortunately.
the devices are all on 9 different subnets. instances of each subnet wake up simultaneously, set their ip address on the respoective device and try to connect to the plc via modbus tcp.
we dont want to keep 200+ instances connected to the gateway.so, logic allows them to wake 9 instances in 9 subnets at a time and try to connect to the plc.
our previous experience suggested that during fail over, the devices cannot be enabled by the redundant partner. the device enable/disable can be done only by master. so we opted for sethostname to be modified.
it seems to be working well when we sequentially connect to these devices.
it only fails when they wake up simultaneously and start changing their device ip addresses.