Since today’s maintenance, 2,000 gateways have been offline. I posted about it in the #ops channel on Slack two hours ago, but there’s been no response.
I’ve tried pinging a contact to make sure they’re aware of the issue. However, please keep in mind there is no SLA for the sandbox environment.
Even without a SLA, this is annoying… Mine is one of those 2000, which means I’m losing the data from the temperature sensors in my room which is of course not a big deal but still not nice since it results in ugly jumps in the monitoring graphs.
And realistically, “Sandbox” (not a fitting name considering probably every single private user is using that) is all that a private individual can afford since 190€/month is crazy expensive. If it was 2€ or even 5€ per month, then sure, I would be happy to pay for better availability etc and to support the service.
2 Likes
You should code for gaps with an interpolation algorithm - and also filter for any reading spikes that are way out of the norm. This keeps the graphs sane and the alerts/alarms to the real deal. Although a ‘no-uplink’ alert is useful.
Perhaps it’s telling you what to expect so you don’t get any ideas about the free thing you get. The more private users, the more it costs to provide the free thing.
Or perhaps it’s a realistic minimum to service an account. It does come with 1,000 devices included which works out at 19c a device a month. With the configuration of organisations, you could rent an instance and sub-let the spare 980 or so devices …
2€ is the minimum you can charge for and still get some change out of card processing and admin costs. And ideally you need to charge per device as the usage is proportional.
You, me and a measurable percentage of other users, but it’s a big leap for anyone, even TTI, to setup a service with all the prior preparation that involves, to service micro-payment customers because regardless of what you pitch, the price point you set & the T’s & C’s, expectations are always higher than what’s being offered and it only takes one customer a day to spend an hour having their (typically self-inflicted) issue resolved to make the whole endeavour moot.
If I could be assured that people would take up a service in the spirit intended and own their own issues, I could start up a TTS instance tomorrow. But 30 years of trading & seeing all the free & freemium offerings falter & die tells me that that is unlikely to occur.
I also saw that my TTIG does not properly work today. It shows “connected” in console, but the logs from the gateway show: MBEDTLS: bad message length and SSL - An invalid SSL record was received. The “fallback” also does not work.
Yep, my TTIG died today as well. Is there a way to get it back online?
It looks like those old devices from 2020 with 2.0.4 firmware are now dead in the water…
How on earth do you come to that conclusion?
Hopefully not dead but currently in a coma! Let’s see what next 24hrs shows….
Cross post of my own observations just posted to Slack #ops:-
Ditto here: seeing the reports earlier on the Forum I checked across my TTN deployed fleet tonight and looks like all recently active TTIGs I’ve looked at so far are offline 
FW 2.0.4 in each case. Even the normally bulletproof ones appear dead. Console show some as ‘Connected’ but closer check shows not seen for ~12hrs - basically around time of TTS update. Only one still physically here at lab/office so tried manual power cycle - no joy. Will leave all to see if they recover through normal 24hr CUPS check cycle but fact nothing happened with forced reset doesn’t leave me hopeful. Main problem with my fleet is many have been deployed over several years and whilst still actively providing TTN Community coverage many are in locations and hosted sites that I can no longer be sure of access to if any manual intervention or reset needed! Others are at sites where they get visited maybe once or twice a year most with no visits planned (many miles away!) for several months to come… I also looked at some other peoples known TTIG’s across various Communities and story the same so assumption its all TTIG’s/specific age of build likely correct. I just checked a ‘similar’ - housing and overall construction + BasicStation - Brown mini-pro deployed over in Florida and that appears working - states connection started ~13hrs ago - so likely around time of last TTS update to nam.
I will dig out a couple of spares in the morning and try to run them up tomorrow as new registrations and see what happens, 
Hi Nick perhaps fire up the TTIG I shipped you back in 1H22 and see what you can discover - interested to see what observations you make or see happen given you likley understand the network exchanges better than I would
Trying to connect my own custom BasicStation gateway using LNS, I can’t get a connection with any one of the ‘minimal set’ of certificates recommended by TTS. So more likely something up with a BasicStation endpoint than the certificates themselves.
My indoor gateway is also offline since yesterday noon, this happened already 4 weeks ago. The MRTG chart pulls the last-seen timestamp from a connected sensor on regular basis and create graphs.
Likely that ties back to last issue (LE crt problem that was rolled back)
Same problem with my 2 TTIG gateways: Since approximately 2026-07-02 12:30 CEST, two TTIG gateways at different geographical locations and different ISPs went offline simultaneously. Both had been operating continuously for several years on The Things Stack Sandbox. LEDs alternate between blinking green and green/red (or green/yellow). Last messages 2026-07-02:12:35 Connect gateway and Receive gateway status, followed by Disconnect gateway. No uplinks or downlinks since then. Hardware reset did not help
I’ve been assured The Things Industries is aware of the issue and working to solve it.
3 Likes
Do I just have bad luck? Removed a gateway from my to add on another account.
Now I am not able to add a the gateway. I am kicked out of the console trying adding the gateway.
Anyway. I will wait.
One of my two TTIG came back today. The other is still offline
What ever was ‘fixed’ looks like around 1/3rd of dropped GW’s are now back, sadly not yet any of mine 
Yep… some more came back earlier this morning, but quite a few still missing.
Edit: Looks like mine actually came back as well earlier today!
Yep, a few hundred more bubbled up this morning inc. a small handful of mine - many still down - another 24hrs and check again?
I can confirm my TTIG is also still offline following the July 2 maintenance. Like many others here, I see around 1000 fewer gateways shown on the official Network Status page compared to before the maintenance.
My device has been stuck in a connection loop. It is frustrating to see that while the status page says ‘Completed,’ a significant number of us are still without service. I am just one more statistic in this large group of affected users hoping for a resolution or an acknowledgment that this is a widespread server-side issue, not an isolated hardware/firmware/SSL failure.