# Gateway.xml documentation and recommendations

**URL:** <https://forum.inductiveautomation.com/t/gateway-xml-documentation-and-recommendations/77032>\
**Category:** Ignition\
**Tags:** ignition79\
**Created:** [July 27, 2023, 6:28pm UTC](https://forum.inductiveautomation.com/t/gateway-xml-documentation-and-recommendations/77032 "2023-07-27T18:28:01Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![salbrechtsen](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/salbrechtsen/32/18660_2.png) [@salbrechtsen](https://forum.inductiveautomation.com/u/salbrechtsen)\
**Post date:** [July 27, 2023, 6:28pm UTC](https://forum.inductiveautomation.com/t/gateway-xml-documentation-and-recommendations/77032/1 "2023-07-27T18:28:01Z")

</div>

Hello!

I'm looking for more info regarding the gateway.xml options in ignition 7.9. Mainly, those surrounding catapult

- acceptCount
- maxThreads
- maxKeepAliveRequests
- keepAliveTimeout
- connectionTimeout

I can't find anything in the 7 documentation. The 8 documentation [here](https://docs.inductiveautomation.com/display/DOC81/Gateway+XML+File+Reference) only has reference to maxThreads, but its listed under gateway not catapult.

---

<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:** [July 27, 2023, 6:32pm UTC](https://forum.inductiveautomation.com/t/gateway-xml-documentation-and-recommendations/77032/2 "2023-07-27T18:32:30Z")

</div>

I think the latter 3 aren't used any more, but `acceptCount` and `maxThreads` are parameters passed to the underlying web server that control how many connections the server socket can have queued up waiting and how many threads it uses to serve requests.

---

<div class="post-metadata">

**Author:** ![salbrechtsen](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/salbrechtsen/32/18660_2.png) [@salbrechtsen](https://forum.inductiveautomation.com/u/salbrechtsen)\
**Post date:** [July 27, 2023, 8:36pm UTC](https://forum.inductiveautomation.com/t/gateway-xml-documentation-and-recommendations/77032/3 "2023-07-27T20:36:31Z")

</div>

Good to know. Is that just gateway webpage like threads, or could it impact client communications with the webserver etc.

What might happen if the acceptCount is exceeded? Would it ever be recommended to increase these values?

---

<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:** [July 27, 2023, 8:40pm UTC](https://forum.inductiveautomation.com/t/gateway-xml-documentation-and-recommendations/77032/4 "2023-07-27T20:40:00Z")

</div>

It includes Vision Client and Designer comms, which also use HTTP(S).

If accept count was exceeded whatever was on the other side would see their connection attempt fail.

Generally, you should not need to increase these values unless you're using hundreds of concurrent clients/designers. You'll probably end up on a support call about your system scaling before you should touch these.

---

<div class="post-metadata">

**Author:** ![salbrechtsen](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/salbrechtsen/32/18660_2.png) [@salbrechtsen](https://forum.inductiveautomation.com/u/salbrechtsen)\
**Post date:** [July 27, 2023, 8:46pm UTC](https://forum.inductiveautomation.com/t/gateway-xml-documentation-and-recommendations/77032/5 "2023-07-27T20:46:51Z")

</div>

Thanks Kevin. We are operating at a pretty large scale and may schedule a support call to discuss.

We had one gateway that was struggling to reboot in reasonable time, when we modified the maxThreads value higher (roughly 2x) we saw the reboot time drop dramatically and the system seems to run better. The gateway in question has a large number of projects, and dozens of designers open usually, so that might explain why it helped.

---

<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:** [July 27, 2023, 8:49pm UTC](https://forum.inductiveautomation.com/t/gateway-xml-documentation-and-recommendations/77032/6 "2023-07-27T20:49:36Z")

</div>

If you take some thread dumps you should see up to as many HTTP/webserver threads as you have configured, and if those threads are constantly in a busy state serving some request rather than just waiting you may very well benefit from more.

---

<div class="post-metadata">

**Author:** ![salbrechtsen](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/salbrechtsen/32/18660_2.png) [@salbrechtsen](https://forum.inductiveautomation.com/u/salbrechtsen)\
**Post date:** [July 27, 2023, 9:00pm UTC](https://forum.inductiveautomation.com/t/gateway-xml-documentation-and-recommendations/77032/7 "2023-07-27T21:00:06Z")

</div>

Thanks. We will check that out.
