# The tag reading on the gateway network takes too long **URL:** \ **Category:** Ignition\ **Tags:** ignition81\ **Created:** [April 7, 2025, 7:17am UTC](https://forum.inductiveautomation.com/t/the-tag-reading-on-the-gateway-network-takes-too-long/104768 "2025-04-07T07:17:31Z")\ **Posts on this page:** 10\ **Page:** 1
**Author:** ![Rui\_Gao](https://avatars.discourse-cdn.com/v4/letter/r/b5e925/32.png) [@Rui\_Gao](https://forum.inductiveautomation.com/u/Rui_Gao)\ **Post date:** [April 7, 2025, 7:17am UTC](https://forum.inductiveautomation.com/t/the-tag-reading-on-the-gateway-network-takes-too-long/104768/1 "2025-04-07T07:17:31Z")
The `references.idb` file located at `\Ignition\data\diagnostics\tags\default\` is experiencing excessive write I/O. Compared to other system tag counts on the gateway, this is affecting the gateway network's tag reading, causing delays of tens of seconds before values can be read. I would like to know which operati ![ignition Wirte IO](https://us1.discourse-cdn.com/inductiveautomation/original/3X/5/4/54c24fad3fb2bc3a48be84f65d08892f92c92897.png) ons or functions might be causing writes to this file. ---
**Author:** ![Rui\_Gao](https://avatars.discourse-cdn.com/v4/letter/r/b5e925/32.png) [@Rui\_Gao](https://forum.inductiveautomation.com/u/Rui_Gao)\ **Post date:** [April 7, 2025, 7:18am UTC](https://forum.inductiveautomation.com/t/the-tag-reading-on-the-gateway-network-takes-too-long/104768/2 "2025-04-07T07:18:56Z")
![normal IO](https://us1.discourse-cdn.com/inductiveautomation/original/3X/4/9/498a59daeaf5eb9b95b35f2ebc389ed8621ef755.png) ---
**Author:** ![pturmel](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/pturmel/32/3295_2.png) [@pturmel](https://forum.inductiveautomation.com/u/pturmel)\ **Post date:** [April 7, 2025, 11:40am UTC](https://forum.inductiveautomation.com/t/the-tag-reading-on-the-gateway-network-takes-too-long/104768/3 "2025-04-07T11:40:29Z")
That is the tag references tracker. If your gateway disk storage is too wimpy to use it with your application, disable it (per tag provider). ---
**Author:** ![paul-griffith](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/paul-griffith/32/37249_2.png) [@paul-griffith](https://forum.inductiveautomation.com/u/paul-griffith)\ **Post date:** [April 7, 2025, 4:03pm UTC](https://forum.inductiveautomation.com/t/the-tag-reading-on-the-gateway-network-takes-too-long/104768/4 "2025-04-07T16:03:06Z")
> [@Rui\_Gao](#): > > gateway network's tag reading I would be surprised if this is actually true, also. You should contact support, but the TRT was pretty carefully engineered to be independent of main tag execution re: threading. So unless the whole system is struggling to keep up, it's unlikely TRT is actually the root cause here; more likely a symptom of some other problem. ---
**Author:** ![pturmel](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/pturmel/32/3295_2.png) [@pturmel](https://forum.inductiveautomation.com/u/pturmel)\ **Post date:** [April 7, 2025, 4:06pm UTC](https://forum.inductiveautomation.com/t/the-tag-reading-on-the-gateway-network-takes-too-long/104768/5 "2025-04-07T16:06:27Z")
> [@paul-griffith](#): > > unlikely TRT is actually the root cause here Is IA testing gateway hardware that **doesn't** have a solid-state disk? (Spinning rust crushes DBs that try hard to not corrupt themselves. SQLite is a terrible choice for the TRT.) ---
**Author:** ![paul-griffith](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/paul-griffith/32/37249_2.png) [@paul-griffith](https://forum.inductiveautomation.com/u/paul-griffith)\ **Post date:** [April 7, 2025, 4:13pm UTC](https://forum.inductiveautomation.com/t/the-tag-reading-on-the-gateway-network-takes-too-long/104768/6 "2025-04-07T16:13:48Z")
I would ~~assume~~ hope so, some pretty careful performance testing was done, but I don't actually know for sure. I also would've thought we'd use SQLite's WAL mode for TRT, though a quick glance doesn't show that we are 🤷. I wasn't involved in the backend, just know from osmosis some of the concerns around implementation. ---
**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:** [April 7, 2025, 4:14pm UTC](https://forum.inductiveautomation.com/t/the-tag-reading-on-the-gateway-network-takes-too-long/104768/7 "2025-04-07T16:14:19Z")
Heh, I would assume not 🤷 and don't even think about using anything but an SSD for Ignition 8.3... ---
**Author:** ![Rui\_Gao](https://avatars.discourse-cdn.com/v4/letter/r/b5e925/32.png) [@Rui\_Gao](https://forum.inductiveautomation.com/u/Rui_Gao)\ **Post date:** [April 8, 2025, 8:30am UTC](https://forum.inductiveautomation.com/t/the-tag-reading-on-the-gateway-network-takes-too-long/104768/8 "2025-04-08T08:30:06Z")
I found the reason, there are multiple WebDevs that receive data and then write tags in batches, when I turn off WebDev, the IO drops a lot obviously, and then the tag read speed of the gateway network also recovers to tens of ms, I improved the method, use the database to store the data, and then write the tags from the database at regular intervals, ---
**Author:** ![Rui\_Gao](https://avatars.discourse-cdn.com/v4/letter/r/b5e925/32.png) [@Rui\_Gao](https://forum.inductiveautomation.com/u/Rui_Gao)\ **Post date:** [April 8, 2025, 8:34am UTC](https://forum.inductiveautomation.com/t/the-tag-reading-on-the-gateway-network-takes-too-long/104768/9 "2025-04-08T08:34:15Z")
![IOChange](https://us1.discourse-cdn.com/inductiveautomation/original/3X/5/f/5f585aa4d5aafda51ecd26360c609289d1fa4db9.png) ---
**Author:** ![Rui\_Gao](https://avatars.discourse-cdn.com/v4/letter/r/b5e925/32.png) [@Rui\_Gao](https://forum.inductiveautomation.com/u/Rui_Gao)\ **Post date:** [April 8, 2025, 8:55am UTC](https://forum.inductiveautomation.com/t/the-tag-reading-on-the-gateway-network-takes-too-long/104768/10 "2025-04-08T08:55:54Z")
![IO-3](https://us1.discourse-cdn.com/inductiveautomation/original/3X/a/5/a582144dd8a694c89fb62c1b3273a78f4640d84b.png) ![NetWork](https://us1.discourse-cdn.com/inductiveautomation/original/3X/a/6/a666c1963b9829f1dcb1d7bdfdbf588235ef4d4c.png) I turned off the switch and then the IO decreased a lot, but the tag reading speed of the gateway network did not improve, it was still very slow, and this suspicion was that there were a large number of write tasks for the tag. Clogged up on read tasks