# Slow systems troubleshooting

**URL:** <https://forum.inductiveautomation.com/t/slow-systems-troubleshooting/25875>\
**Category:** Ignition\
**Tags:** ignition79\
**Created:** [June 17, 2019, 2:56pm UTC](https://forum.inductiveautomation.com/t/slow-systems-troubleshooting/25875 "2019-06-17T14:56:14Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![apimente](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/apimente/32/4030_2.png) [@apimente](https://forum.inductiveautomation.com/u/apimente)\
**Post date:** [June 17, 2019, 2:56pm UTC](https://forum.inductiveautomation.com/t/slow-systems-troubleshooting/25875/1 "2019-06-17T14:56:14Z")

</div>

Friends,

I’m sorry for this “open” question but i need some troubleshooting tips for slow response system. This is a frecuently fact, but this time is about ignition 8, kepserver 6 and 3 plc CLX 73 with 1756-EN2T ethernet gateway.

For now, I know that we have almost 21000 tags (that’s what KS says) divided in 3 PLC. Here you can see the diagnostic page:

 ![image](https://us1.discourse-cdn.com/inductiveautomation/original/2X/b/b114c7fbbc6912ddb74e68dc7c75ff8f99b1b939.png)  
We expect that TAGS with 500ms must respond to changes in that order of time.

Any help and hints are welcome. Greetings,

---

<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:** [June 17, 2019, 3:10pm UTC](https://forum.inductiveautomation.com/t/slow-systems-troubleshooting/25875/2 "2019-06-17T15:10:43Z")

</div>

If these tags are all from Kepserver you should be looking there for troubleshooting.

[Their manual](https://www.kepware.com/getattachment/c7cd42b2-e3a8-403e-ac5b-ddff9cb03985/allen-bradley-controllogix-ethernet-manual.pdf) has a section on performance optimizations, have you looked at that yet?

---

<div class="post-metadata">

**Author:** ![apimente](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/apimente/32/4030_2.png) [@apimente](https://forum.inductiveautomation.com/u/apimente)\
**Post date:** [June 17, 2019, 7:25pm UTC](https://forum.inductiveautomation.com/t/slow-systems-troubleshooting/25875/3 "2019-06-17T19:25:39Z")

</div>

Thanks Kevin. There are some hints that we can follow. But in Ignition, do you know something about the interpretation of the info that I attached?

For example, for “direct\_1000” scanclass, We have 5700 tags so the average duration should be 1000/5700 [ms] or there are concurrent threads inside the scanclass or another kind of optimization.

Thanks again for the KS manual.

---

<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:** [June 17, 2019, 7:28pm UTC](https://forum.inductiveautomation.com/t/slow-systems-troubleshooting/25875/4 "2019-06-17T19:28:59Z")

</div>

“Scan class” is a misnomer when it comes to OPC tags. All we do is pass the scan class rate along when creating a subscription and monitored items on the OPC server. There is no polling or scanning happening between OPC client and OPC server.