# ProxySQL exporter returns odd values

**URL:** https://forums.percona.com/t/proxysql-exporter-returns-odd-values/6293
**Category:** PMM 1.x
**Created:** [April 10, 2018, 4:47am UTC](https://forums.percona.com/t/proxysql-exporter-returns-odd-values/6293 "2018-04-10T04:47:23Z")
**Posts on this page:** 2
**Page:** 1

<div class="post-metadata">

### Author: ![DmytroKh](https://avatars.discourse-cdn.com/v4/letter/d/e79b87/32.png) [@DmytroKh](https://forums.percona.com/u/DmytroKh)
#### Post date: [April 10, 2018, 4:47am UTC](https://forums.percona.com/t/proxysql-exporter-returns-odd-values/6293/1 "2018-04-10T04:47:23Z")

</div>

Hi,  
I’ve tried to setup dashboard to get an email once any of slave servers is out on proxysql pool  
however it appears proxysql\_connection\_pool\_status returns values other than 2,3,4

Kindly see prometheus/graph attached,  
the metric returns 1 for ONLINE, but once server got SHUNNED , on reached max\_replication\_lag treshold, the metric returns some 3-digit values, that actually looks like some random values .

 ![photoid=51284](https://us1.discourse-cdn.com/flex019/uploads/percona1/original/2X/e/e1838720f2b4f15560f73923400c0e7b9b425c54.png)

---

<div class="post-metadata">

### Author: ![DmytroKh](https://avatars.discourse-cdn.com/v4/letter/d/e79b87/32.png) [@DmytroKh](https://forums.percona.com/u/DmytroKh)
#### Post date: [April 10, 2018, 5:31am UTC](https://forums.percona.com/t/proxysql-exporter-returns-odd-values/6293/2 "2018-04-10T05:31:56Z")

</div>

In addition,

As I see “Proxysql overview” dashboard works fine and shows status correctly, However graph shows weird values, I’ve set an alert in Grafana based on graph and see that random values on emails and alert history as well

 ![snapshot6.png](https://us1.discourse-cdn.com/flex019/uploads/percona1/original/2X/c/cd27cc6ccf83aa3b7483c6f14c9adee74a4d96ce.png)