Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi
We use Qlik Sense Enterprise Feb 2020 version in a 8-node cluster. Apps are loadbalanced to dedicated customer-facing nodes.
Recently (after upgrading to Feb 2020), we found often the lab between QMC & hub & mashups.
For example, changing name of data connection takes hours to be visible in hub. I changed owner of a private sheet couple of days ago, the sheet is visible to me today. Colleague tried to add label to dimension using API, it appeared a day after.
We don't have RAM/CPU issue when lagging happened.
Do you have any suggestion where to check? Our dedicated postgresql server is fairly small, very little CPU & RAM usage. Is it something we should look into?
Best regards,
Susan
This morning, I changed postgresql max_connection to 990. (We added 2 nodes a week ago, but I forgot to change max_connection).
My test after increase max_connetion:
1. Edit a data connection in hub. The data connection has my name next to it.
2. I go to QMC to remove my name from name of the data conneciton.
3. Come back to hub, refresh page, my name is still there.
Refresh regularly, after 30min, my name is still there...
Hi,
I have the same problem. Have you found an answer?
We had experienced this issue for ages, never knowing why it happened. We inspected the traffic and it could not be a proxy cache because the information about data connections in the Data Load Editor are passed via Websocket.
After trying many things, we finally managed to reproduce the issue. It is due to having two data load editors open at the same time in different tabs and then changing the name in the QMC. The issue has been reported to Qlik with support ticket 00147307.