Duplicate Tag IDs After Redundancy Failover

Ignition
1

Hi everyone,

I’m experiencing an issue with Ignition 8.1.45 (upgraded from 8.1.7 with the same problem) running in a redundant setup connected to a Postgres 13 database. Here’s the scenario:

  1. When new history tags are created while the first gateway is in the Master role, the corresponding entries are correctly inserted into the sqlth_te table.
  2. After a failover and subsequent role reversal, when the first gateway becomes the Master again, Tag Historian re-creates the IDs of the most recently created tags. However, it reuses IDs that are already associated with other tags or retire them ¿?.
  3. This results in trends, tables, and reports showing a mix of data from different tags, as queries are executed using the duplicated IDs.

Has anyone encountered this issue before or have any recommendations on how to resolve it? Any insights would be greatly appreciated!

Thanks in advance!

Example 1

Results_1
Results_1676×503 94.8 KB

Example 2:

Results_2
Results_2791×515 192 KB

2

Hello Jorge ,

I am having the same issue, I was wondering if you found the way to fixed?

Best regards.

3

Hi Fernando,

I solved it by stopping both gateways, fixing and sorting the database data manually (I did a tagid map with corrections to be done first), cleaning all store and forward queue by deleting corresponding files in both primary and redundant gateways and starting both gateways again. It seems the problem was originally created when during a failover, there was a queue of the store and forward in the backup server with new tags configured recently to use tag history. It is a very odd behavior and I have seen it only in this particular couple of master-backup configuration.

4

I had the same issue with tags having weird behaviors after failover.

I solved it using the parameter ‘Use Active Time To Resolve Conflicts‘= True.

If that one is = false, on a failover when the Master take the lead again all the new data processed for the backup it will be dropped in favor of the old master data(all the data on the redundancy provider).

if that one is = true (I changed the set up in both gateways) all the data processed when the back was active it is going to be pushed to the master when that one takes the control again and avoiding duplicating entries or tags overwritten.

1 Like
5

Very nice! Since I started working with these gateways on version 8.1.7, I didn’t see that option, which is available since version 8.1.31. I will set it up following your solution. Thanks!