I set up a new AirGradient One indoors last week and connected it to the AirGradient dashboard. The monitor itself has been working great, but in two days I’ve gotten 27 email notifications telling me that it’s gone offline:
The status LED is purple, which I understand means the monitor can reach the AirGradient servers but not actually talk to them: Understanding the LED Indicators on Your AirGradient ONE
Is this expected or is there some sort of outage?
Hi @Lantioch,
Thanks for joining! I would be happy to look into this for you if you can reach out to us at support@airgradient.com. If you can provide the serial number of your device, we can look into the root cause of this.
If you would like to disable the notifications, you can do this on the dashboard under ‘Users & Permissions’. Just untick the box for ‘email notifications’ under Actions → Edit.
I sent an email. I was also able to turn off the email notifications. Thanks!
I haven’t been seeing any outage of the Dashboard.
Likely, your Wifi signal is low and is causing the AG to not be able to talk to the servers. In the Dashboard, you can click the 3 lines icon in the upper right and enable “Signal Strength” to get a new column "Signal (dBm) and see what your device is reporting. -70 or lower is a bad signal. I have some at -45 which is great, and some at -75 which work for me but isn’t fantastic. Another user was having disconnects with -79. So that gives you some numbers to compare your device to
Thanks for the suggestion. It’s not a signal issue - the monitor is about 3 meters from the access point, and I am able to ping it without issues. The last time it was up, the signal was -38 dBm:
I had the AirGradient One on an IoT-only VLAN that could be reached from the rest of my network but that could only reach out to the internet, not to the rest of the LAN. Disabling that and turning it back on fixed it, the monitor is back online.
I’ve had this same issue for months, but only with my Open Air unit. After months of being stable, it suddenly started going offline for periods. Some days the monitor is online, other days it’s offline. I spent ages trying to figure out what was wrong. I could never figure out what was causing it, so I just live with it.
I’m actually having this issue too. I reinstalled the firmware (which rolled me back to 3.3.9) and still am stuck with the purple light, but also cannot access my airgradient one via my dashboard. wifi signal should be good, as I never had issues before and this suddenly cropped up about a month ago. any ideas @Ethan_AirGradient ?
Doing anything special like an isolated VLAN for iot devices, or firewall rules? DNS ad blocking like Pihole or AdGuard?
Firmware 3.3.9 is the latest available on the GitHub site, so I don’t see that as rolling back, as that should be the version pushed.
Does it not connect at all to the Dashboard site, or just intermittently? If it reports sometimes, you can click the 3 boxes icon to add other columns and add the Signal Strength column. -75 or less indicates a very poor signal.
You could also try moving it closer to your AP for awhile and see if anything changes.
Worst case, you can connect a USB cable to it and from the Firmware flashing page you can view the logs so you can see exactly what it is seeing and may give some additional information.
You can also open a support case and include your serial number when reporting it here:
airgradient.com/support/
Hey all, thank you for your replies 
And, sorry, I meant that it rolled me back to 3.3.8 from 3.3.9 (using the web-based flasher)
I have nothing interesting going on with my network like VLAN. I do have NextDNS but parsed through the logs and did not see anything from that device that was blocked. Also, like I mentioned before, nothing has changed about my setup– my AG One worked for several months and then stopped.
From the firmware flashing page, the only logs it would give me were about current sensor data. I suppose I will submit a ticket. I attempted it earlier today but it gave me an error and said to try again later
Appreciate all your help!
Just chiming in here on a related issue. I have an open support ticket (as of Aug 27) to resolve what appears to be a major FW bug. Specifically, I’m experiencing persistent and intermittent dropouts “Unavailable” several times per hour on two brand new AG units: the AG One and AG Open Air (only one week old!).
The drops occur randomly and affect each device at different intervals and last for about 30 seconds each time.
Important to note, this issue appears to be isolated to the Home Assistant Native Integration only! All AG Home Assistant device entities simultaneously are unavailable during a periodic dropout, rendering any type of associated HA script, automation, etc. as unreliable / unusable!
The AG Cloud Dashboard and on-device OLED display metrics are unaffected.
In its current state, it’s a terrible user experience. The so called Air Gradient “Platinum” rating on Home Assistant partner integrations was a key driver from my purchase decision. How on earth did the 3.3.9 release pass QA/QC?
I’m on the same firmware version with my devices I’ve had for quite a while and I’m not seeing dropouts with the HA Integration. What exactly are you seeing? Does HA show it offline, or do graphs not update with new information?
Any special vlan configurations or firewall rules between AirGradients and HA?
I’ve attached a screenshot of a typical intermittent dropout lasting about ~30s, multiple times per hour. I also used Gemini Pro to do deep research into this issue and it found what I also suspect is a “Update Fail Reboot Loop”, since I’m also seeing the Firmware Turn Off and Firmware Turn On activity sequence in the logs after these drop outs. Several corroborating screenshots were submitted to AG support last week. Ticket is open and active (I believe @Ethan_AirGradient is looking into this).
No special Vlan network configuration or firewall rules. I disabled "AD Guard"and assigned static IP’s, with no effect or resolution. For additional context, my HA setup is very large (device count is 340+) and these are the only two devices on the entire network behaving this way,
Thanks for your input and help!
Hi, @MallocArray any further insights on what maybe the root cause of these persistent HA dropouts?
I’m also considering moving over to the ESPHome integration, but I want to give the native integration a solid chance first, considering AG’s recent “Works with Home Assistant” certification and Platinum rating for partner integrations.
No idea why my setup would be an edge case.
I’ve sent device logs to AG for review.
@Altair_AirGradient is assisting.
I would find the IP of your device and then setup a continuous ping from your computer to see if it is truly going offline at all.
You can also go to the Dashboard’s Hardware page:
https://app.airgradient.com/settings/hardware
and see what the “Current Uptime Count” column shows for the device. If it is a low number, that would also confirm that your device is rebooting.
Once you know if it is the device itself rebooting or just Home Assistant losing connection, that will help troubleshooting.
You could also try going to an earlier firmware if you think that is the issue, although I’m running the latest HA and AG firmware and not experiencing the same issues.
https://app.airgradient.com/settings/place?tab=2
From here you can set which firmware you want your device to run. I haven’t done a downgrade this way, but it should work, or you can flash it manually.
@MallocArray Here’s what I’m seeing in terms of uptime, with both devices setup one day apart.
The devices are not rebooting at all as far I I could tell. Pretty stable!
The AG One is in a prominent main floor location and when I look at my HA dashboard while a dropout occurs the device OLED display metrics are unaffected. I also check the AG dashboard. Also not affected when HA dropout occurs.
Note both the AG one and Open Air devices experience HA connectivity issues at different intervals separately.
Thanks again!
Yep, so doesn’t look to be the device rebooting, which is good.
Just for fun, on the Dashboard page, you can click the three bars button on the far left and enable the “Signal Strength” column and see what values you are seeing. If less than -75, the signal is weak and can cause interruptions, so something else to check.
Your Home Assistant is on a recent release?
@MallocArray, here you go…
Signal on both devices are pretty solid! The network infrastructure is also on a pretty beastly setup with an Asus RT-AX89X and 3 hard wired Asus RT-AX86U AP’s (AP’s are running on a MoCa 2.0 network via Ethernet backhaul). Additionally, I have direct Fibre Optic to my home with symmetrical internet 1.5GB Up/Down.
My HA device count is 341 with no notable network issues except for these 2 AG products, which not be happening.
Also, yes I’m pretty good at HA house keeping. On the latest 2025.8.3.
I’m curious what the AG engineering team will discover on the device logs I sent over to them last Friday.
Overall, I have to admit it’s been a pretty frustrating out of the box HA experience last 2 weeks, with my very first purchase of a “Works with Home Assistant” certified product, and Platinum rated!
That said, I really appreciate and grateful for your support and guidance on this issue. I read good things about AG and I want this to succeed!
Yep, you’ve done all the right stuff and confirmed all of the standard things. It does seem to be a narrow case, as I’m not aware of others reporting similar issues. I’m on a similar setup with Unifi APs and HA 2025.8.2 so only one minor patch away.
You would be surprised how many issues have been someone with a device just on the edge of wifi connection, or with a complicated isolated VLAN that doesn’t come into the conversation until later, so I just wanted to check.
At one time there was a bug where if you were using something like prometheus/graphana to pull metrics from the local endpoint it could cause the devices to act up, but that doesn’t seem to be the case here either.
Hopefully the support team gets back to you with some good information soon and you can share the resolution.
I’m not a fan of using multiple APs in the house that don’t coordinate channel usage between one another. It sounds like @MarKusTech is using Asus devices and I believe that these don’t coordinate channels between the APs but rather, the master picks channels and the other APs use the same ones. Contrast this to Ubiquiti where there is a central UniFi control plane that conducts site surveys on all of the APs and assigns channels to each AP so that they minimize interference with one another and with other third-party APs in the local area. IMHO, it’s a more robust system.
I, too, have had some drop-outs on my new AG outdoor device. Strangely, I could not get it to reconnect until I rebooted it several times. This is concerning.
I did run some ping tests and noticed a fair amount of packet loss, indicative of poor signal out in the backyard where the AG is:
24902 packets transmitted, 21461 packets received, +1 duplicates, 13.8% packet loss
round-trip min/avg/max/stddev = 4.089/157.088/3059.163/197.515 ms
So, to eliminate poor WiFi, I ordered a 24VDC industrial ethernet switch and I’m wiring it up to the 24V bus that I have on my backyard weather tower (fed by a landscape lighting controller). I already have direct burial Ethernet going across the lawn to the tower, so I bought another Ubiquiti AP, the outdoor-rated “Swiss Army Knife” with an omnidirectional antenna. Hopefully this improves the situation for the AG.