Tag Clash Cannot Start Ignition Service

Ignition
1

I am running version 8.0.15 and I cannot start my Ignition service anymore. When i check the logs all i see in the wrapper is this message repeating over and over again.

INFO   | jvm 1    | 2020/09/10 14:10:46 | W [t.e.core                      ] [21:10:46]: Tag clash at path '[messagingTags]publishers/masterDataBrand/publisherId', a different tag with the same name already exists. 
INFO   | jvm 1    | 2020/09/10 14:10:46 | W [t.e.core                      ] [21:10:46]: Tag clash at path '[messagingTags]publishers/masterDataBrand/publisherId', a different tag with the same name already exists. 
INFO   | jvm 1    | 2020/09/10 14:10:46 | W [t.e.core                      ] [21:10:46]: Tag clash at path '[messagingTags]publishers/masterDataBrand/publisherId', a different tag with the same name already exists. 
INFO   | jvm 1    | 2020/09/10 14:10:46 | W [t.e.core                      ] [21:10:46]: Tag clash at path '[messagingTags]publishers/masterDataBrand/publisherId', a different tag with the same name already exists. 
INFO   | jvm 1    | 2020/09/10 14:10:46 | W [t.e.core                      ] [21:10:46]: Tag clash at path '[messagingTags]publishers/masterDataBrand/publisherId', a different tag with the same name already exists. 
INFO   | jvm 1    | 2020/09/10 14:10:46 | W [t.e.core                      ] [21:10:46]: Tag clash at path '[messagingTags]publishers/masterDataBrand/publisherId', a different tag with the same name already exists. 
INFO   | jvm 1    | 2020/09/10 14:10:46 | W [t.e.core                      ] [21:10:46]: Tag clash at path '[messagingTags]publishers/masterDataBrand/publisherId', a different tag with the same name already exists. 
INFO   | jvm 1    | 2020/09/10 14:10:46 | W [t.e.core                      ] [21:10:46]: Tag clash at path '[messagingTags]publishers/masterDataBrand/publisherId', a different tag with the same name already exists. 
INFO   | jvm 1    | 2020/09/10 14:10:46 | W [t.e.core                      ] [21:10:46]: Tag clash at path '[messagingTags]publishers/masterDataBrand/publisherId', a different tag with the same name already exists. 
INFO   | jvm 1    | 2020/09/10 14:10:46 | W [t.e.core                      ] [21:10:46]: Tag clash at path '[messagingTags]publishers/masterDataBrand/publisherId', a different tag with the same name already exists.
1 Like
2

I am now on 8.0.16 and still have this problem. I was able to track down how exactly this is happening. If delete and recreate a tag over and over again it will keep a record in the internal database for each of the old version of the tags that was deleted. If I then restart the Ignition service it will get stuck with the above posted errors in the wrapper.

Here is a test you can try that will reproduce the problem.

tagPath = "[tagProvider]testFolder"
tagName = "testStringTag"
tag = {
	"name": "testStringTag",
	"valueSource": "memory",
	"dataType": "String",
	"tagType": "AtomicTag",
	"value": "testValue"
}



for x in range(0,10):
	system.tag.deleteTags([''.join([tagPath, "/", tagName])])
	system.tag.configure(basePath=tagPath, tags=[tag])

my tag tree looks like this after running the code

image

But my internal database table looks like this

image
image836Ă—219 9.15 KB

3

I need to bump this. We have this exact problem on an active production facility. The only solution we found was to uninstall the gateway from the pc and reinstall it. We need a better fix.

4

Have you contacted support?

5

I did just now. Hopefully we can find a solution to this.

6

FWIW, this issue may already be fixed in 8.1.2.

If you can’t upgrade I think there’s also some discussion about backporting this to the 8.0.18 “nightly” release that we’re still doing for some reason.

1 Like
7

Thank you for the info. I’ll try to install the latest version. Hopefully it fixes it.
The latest version currently available is 8.1.1. I’ll install it and see what happen.

8

The ticket I found that looks it’s related to this is for 8.1.2 so maybe don’t get your hopes up that 8.1.1 will fix it.

9

I had this same issue, except my hourly gateway backups had the same issue when restored (they only go back a week). I was able to find a project from September that I could run but unfortunately there’s been changes between now and then that are lost unless I can get one of my backups to work.
I’m not sure if this detail is relevant, but when I was first told that there was an issue, I couldn’t RDP into the server, I had to reboot the VM first.
I was running 8.1.0rc initially, when I reinstalled, I went to 8.1.2.
(I do have a support ticket I just opened for this)