# Historian Tag Group in Driven Mode Occasionally Missing One-Second Sample

**URL:** <https://forum.inductiveautomation.com/t/historian-tag-group-in-driven-mode-occasionally-missing-one-second-sample/117009>\
**Category:** Ignition\
**Tags:** database, ignition81, perspective\
**Created:** [September 7, 2026, 11:29am UTC](https://forum.inductiveautomation.com/t/historian-tag-group-in-driven-mode-occasionally-missing-one-second-sample/117009 "2026-09-07T11:29:13Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![pramod\_thorat1](https://avatars.discourse-cdn.com/v4/letter/p/b38774/32.png) [@pramod\_thorat1](https://forum.inductiveautomation.com/u/pramod_thorat1)\
**Post date:** [September 7, 2026, 11:29am UTC](https://forum.inductiveautomation.com/t/historian-tag-group-in-driven-mode-occasionally-missing-one-second-sample/117009/1 "2026-09-07T11:29:13Z")

</div>

Currently, I am facing an intermittent issue with **Ignition Tag Historian**. I have configured a **Driven Tag Group** with a Boolean trigger tag that remains `TRUE` while the machine is running. The historian is configured to record data at **1-second intervals**.

Most of the time, the data is logged correctly every second, but occasionally **one second of data is missed** , even though the trigger remains continuously `TRUE`.

For example:

```python
16:27:36
16:27:37
16:27:38
16:27:39
16:27:40
16:27:42
16:27:43

```

Here, the `16:27:41` record is missing.

I would like to understand why this happens and whether a **Driven Tag Group can occasionally skip an execution/sample** due to Gateway load, tag execution timing, OPC communication, historian processing, or database delays.

My requirement is to reliably record **one historian sample every second without missing any samples** while the machine is running.

I have attached a screenshot showing the normal 1-second historian data logging for reference.

 ![image](https://us1.discourse-cdn.com/inductiveautomation/original/3X/3/1/31d83ae82a572d086df38ff78a08cdd327ee96e4.jpeg)

---

<div class="post-metadata">

**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:** [September 7, 2026, 1:02pm UTC](https://forum.inductiveautomation.com/t/historian-tag-group-in-driven-mode-occasionally-missing-one-second-sample/117009/2 "2026-09-07T13:02:24Z")

</div>

> [@pramod\_thorat1](#):
>
> can occasionally skip

Yes. For any of the reasons you list. Possibly also GC stalls.

> [@pramod\_thorat1](#):
>
> My requirement is to reliably record **one historian sample every second**

The only reliable way to guarantee this kind of recording is to use a buffer in the PLC, and use a pair of handshake tags to signal data available and acknowledge data stored to DB, per record. Usually combined with a scripted OPC read to ensure data is collected **after** the signal. (Ignition tags cannot actually do this part--they are entirely asynchronous to anything else, and OPC subscriptions do not guarantee any kind of delivery order.)