# OPCUA Subscription Sychronizer - Failed to Create Subscription

**URL:** <https://forum.inductiveautomation.com/t/opcua-subscription-sychronizer-failed-to-create-subscription/94203>\
**Category:** Ignition\
**Created:** [August 6, 2024, 12:32am UTC](https://forum.inductiveautomation.com/t/opcua-subscription-sychronizer-failed-to-create-subscription/94203 "2024-08-06T00:32:26Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![Thomas\_Magill](https://avatars.discourse-cdn.com/v4/letter/t/65b543/32.png) [@Thomas\_Magill](https://forum.inductiveautomation.com/u/Thomas_Magill)\
**Post date:** [August 6, 2024, 12:32am UTC](https://forum.inductiveautomation.com/t/opcua-subscription-sychronizer-failed-to-create-subscription/94203/1 "2024-08-06T00:32:27Z")

</div>

We have a couple OPC UA devices on our network that drop connection consistently around the same timeframe, every day, which happens to be when they would be expected to pass the most data. The errors I am seeing are OpcUaSubscriptionSynchronizer "Failed to create subscription 'tag-group\_default", and DataTypeTree Browsing failed.

 ![ignitionErrorScreencaps](https://us1.discourse-cdn.com/inductiveautomation/original/3X/3/0/304f75cc4a55265dfac307e969fa494c9d8c6528.png)  
[IgnitionLogsOPCUA.zip](https://forum.inductiveautomation.com/uploads/short-url/6Os9KLydcLIYUZbIs294GKnuLFA.zip) (2.0 MB)  
I set the logger to debug and attached.  
I've seen a couple potential solutions on here, such as raising Max Per Operation, or settings the tags to a different tag group altogether (we have 500k tags).  
Thoughts?

---

<div class="post-metadata">

**Author:** ![Kevin.Herron](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/kevin.herron/32/7259_2.png) [@Kevin.Herron](https://forum.inductiveautomation.com/u/Kevin.Herron)\
**Post date:** [August 6, 2024, 12:55am UTC](https://forum.inductiveautomation.com/t/opcua-subscription-sychronizer-failed-to-create-subscription/94203/2 "2024-08-06T00:55:58Z")

</div>

You could try _lowering_ the Max Per Operation for the problematic connections as well as increasing the Keep Alive Allowed and Timeout settings.

The problem you're running into is that these servers go unresponsive for some reason, Ignition's client eventually decides to reconnect because the server isn't responding, but then unsurprisingly the server is still not responsive to the things that need to happen upon reconnecting.

---

<div class="post-metadata">

**Author:** ![Kevin.Herron](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/kevin.herron/32/7259_2.png) [@Kevin.Herron](https://forum.inductiveautomation.com/u/Kevin.Herron)\
**Post date:** [August 6, 2024, 12:56am UTC](https://forum.inductiveautomation.com/t/opcua-subscription-sychronizer-failed-to-create-subscription/94203/3 "2024-08-06T00:56:49Z")

</div>

> [@Thomas\_Magill](#):
>
> or settings the tags to a different tag group altogether (we have 500k tags).

Unless you're talking about 500k tags from one server this probably isn't necessary. How many servers, how many tags per problematic server?

---

<div class="post-metadata">

**Author:** ![Thomas\_Magill](https://avatars.discourse-cdn.com/v4/letter/t/65b543/32.png) [@Thomas\_Magill](https://forum.inductiveautomation.com/u/Thomas_Magill)\
**Post date:** [August 6, 2024, 1:26am UTC](https://forum.inductiveautomation.com/t/opcua-subscription-sychronizer-failed-to-create-subscription/94203/4 "2024-08-06T01:26:04Z")

</div>

Its only two devices, and they're only serving up a couple tags each. We have ~40 of these devices on the network and its the same two causing issues consistently.  
Internal logs on the devices themselves also show "network" related issues, and the ethernet runs are close to 300ft long so we're looking into shortening the run.