TTIG and The Things Stack

Soon after The Things Conference, we will host a TTI CUPS server to which all TTIGs will be migrated. After that you can configure your gateway to any server you want. So for now, TTIGs will remain on TTN v2.

3 Likes

Thanks for update and confirming

1 Like

I deleted … but he was used for test purpose no worry. After the conference, I will use TTI cups…
Many thanks.
Fred

1 Like

I should say give it a try again, and try to reconnect on the old way. You should see very quick if the ID still is in use.

Pro Tip: Since v2 is already connected to Packet Broker, traffic received by your TTIG will be forwarded via Packet Broker to TTN V3. This is already setup

  • Simply migrate your device to v3 and leave the gateway as it is. Your devices will join via your TTIG which is connected to V2 (through Packet Broker).

You can use that flow until we have the TTI CUPS ready for you.

Does this include the Browan Minihubs not labelled as TTIG but linked to TTN Stack v2 or is this a task the distributors must do?

Yes this is included.

1 Like

Was a bit quick late last night and deleteing my TTIG on v2 and tried to add it to v3 - whiteout success. Yes I saw the warning, but just thought “… not going to connect it to v2 again anyway …”. Is there any possibility the get it back on to v2? Or v3 at a later state? Generate a random EUI?

Is there any possibility the get it back on to v2

Unfortunately not.

Or v3 at a later state? Generate a random EUI?

A production gateway EUI is fixed.

I can work some magic for you. Write me a direct message.

1 Like

Hello I would like to know how we can make a TTIG-868 target TTS (locally) or another server instead of the initial configuration? I would like to test a C class of a device but TTNV3 doesn’t seem to offer it (like V2)?

Search the forum for the command line instructions to ‘claim’ the gateway or use the help included in the commandline client. Use the TTIG WiFi password as claim code.

If you add a device manually to V3 you have the option to select class C. May-be you need to take a second look?

1 Like

Thank you yes I saw this option but I do not have the impression that it is active, in particular in the eu1 cloud community version.
I also find that a lot of options are missing in the V3 compared to the V2. Example the f_cnt uplink / downlink reset, integrations … Are there any changes planned?

I think you need to get more familiar with V3. There is a lot available in different places, however the stack is much more mature and offers more options.

Class C is available in the community stack and if you search the forum you can find messages by members using it.

For reset of the counters, because V3 has a state for each device you need to reset that. At the moment only available using the command line tool but if you log a ticket at GitHub or find the forum topic where TTI was soliciting feedback you might see it in the web console as well in the future.

Ok I thought the V3 would be more intuitive to use! If I have or a user has to spend a lot of time doing (thorough) research to find out where features are and how to use them it’s boring.
I have an example, in V2 it was easy for the same End Device to switch from ABP mode to OTAA mode and vice versa. There I had to remove the End Device in ABP to recreate one in OTAA (the radiochecks were grayed out). And there are plenty of other examples.

Is it possible to create an http POST request like in V2 (via HTML integration)? I can’t find a link, I typed TTN V3 integration html post
https://www.thethingsnetwork.org/docs/applications/integrations/http/index.md

best regards

You mean click on a few menus to see what the options are? If you know LoRaWAN, it’s like getting behind the wheel of a different vehicle - same controls, just in different places. Unless like me you have a attention span of a, ooooooh, squirrel :chipmunk: :chipmunk: :chipmunk:, unmedicated ADDer - where upon you find out that your super hero special power is the ability to absorb data at 10x NT bandwidth (NT = neurotypical aka normal people).

TTS is the grown up version to v2’s finding it’s way. I create an ABP device for initial testing and then create an OTAA device for final testing. Not hard, takes 2s using the CLI.

Do share your other examples so we can add it to the gotcha’s list.

For real?? You think that may have been deprecated?

Suggest you take time out to find out which side of the steering wheel the indicators are aka find the v3 docs (hint, linked to on every page of the console).

And for the bad news, the payload / json format has changed. But if you search the forum, you’ll find the free scripts to get you started as provided by someone awesome who read the docs 9 months ago.

1 Like

I do not question your abilities I know you are good and active on the network and that you are a great person but rather than playing guesswork with literary answers, the easiest way is to indicate the links to retrieve the information as quickly as possible. I just don’t have time to read the details. There are intuitive parts but frankly take a reading as for the V2 I do not have that to do. I am a solution designer and national trainer and frankly the V3 version is not complete. A user does not have time to spend on where to find information in the forums. YourI never said that switching from ABP mode to OTAA mode in the TTS V3 was complicated but on the other hand it is worse than on the V2. Because the concern stems in particular from the fact that we will no longer be able to give the same name if we go back to ABP mode again. CLI tool should appear directly in an item etc …

Regarding the POST request in HTML, it is not the problem or not that the method is obsolete, it is to show the feasibility. Don’t worry I created a template-less webhook for this. But that’s not intuitive again.
Another example ? Yes of course, we do not see the integrations that we have created (nor if it is running) in relation to an application. It’s a shame not to have an overview of the whole without having to click on the type of integration.

Change is hard and people resist change. It requires you, like everyone else, to invest time to get to know the new stuff. I expect you are used to that as a trainer as what you teach will change constantly as well.

The documentation you search is available from the links at the bottom of every page in the console, no digging in the forum required.

1 Like

Sure, I have people that I provide the links for those who don’t have time to read. I call them A list clients. They get invoices, fairly substantive ones. As there is a limit to how many I can do this for, there are very few of these people and it is only done for a finite amount of time. So in parallel their staff are trained, for whom I also provide links / cheat sheets and various other resources and, of course, invoices. If after a reasonably timescale I’m not relegated to third tier support, then the invoices get painful until I am, because I am a very strong believer in CPD seeings as how so many people have demonstrated that they would rather piggy back off other peoples hard work.

So my default position on the forum is to “teach a man to fish” rather than just spoon feed. If only so they at least try to find answers before falling back on tapping in to the collective knowledge base.

That said, I’m happy to create resources (like the integrations templates, device creation API scripting, the CLI command tree, the EUI generator) to fill in some of the gaps and put them out for public consumption, so it’s not all about the invoicing.

One of my many other foibles also runs to supporting the awesome gift that TTI provide us with. Arguements range back & forth about how the TTN community has benefited & how TTI has benefited, but at the end of the day, we are given access to software & hardware for free. It seems a bit churlish to complain that we also don’t have a personal concierge service to answer questions on demand because we are too busy using the free resources provided to take time out to learn the new version.

Between your TTIG claiming thread and now this thread, it’s clear you can save yourself a bunch of time by scanning through the documentation - read all the opening paragraphs on every page so you will have a rough idea of where to look going forward.

2 Likes

I am a solution designer, an international consultant/developer/trainer/international-man-of-mystery and I’ve been using TTS live on both CE and my own hosted instance for many months and apart from downlink confirmation events which only arrived last week, haven’t found anything missing.

This is the sort of problem I like to solve - I’ll be kind and assume you mean ID and not name, but you can delete & re-create a device with the same ID but with a different set of settings. I can script the change easily enough with the CLI. Not as simple as v2, but TTI would rather we all use OTAA.

I’m not sure I understand what’s missing here - you can enter a URL and set all sorts of path options - perhaps this workshop video from the TTS mini-conference will help.

The only one you can have more than one of is Webhooks. MQTT you can’t turn on & off but it would be useful to know what API keys have been setup that allow access to the feed. Storage Integration you can only turn on or off, no other settings, but same with the API keys. AWS IoT & LoRa Cloud you can setup, again, same with keys.

Whilst I haven’t got that far in porting my management database application to cover v3 integrations, that sort of overview I’d create using the API. But mostly the configuration documentation that the SysOps have & the access controls to prevent anyone dipping in to integrations without prior planning, change request etc, means that having an overview list like we had in v2 rather academic. I think once a project is live that it is very rare of anyone to go in to that section, so I’ll not be stampeding to do anything in that area.

I’m still trying to figure out if a sort of password manager for the API keys is a good idea. It seems appropriate to me that TTS doesn’t show the key after you’ve got it setup, but I’ve got a whole list of sundry keys now and as suggested above, it’s hard to know which ones allow what without looking at each in turn. So maybe not a password manager (although it is useful to have somewhere to record them other than embedded in a settings file somwhere), but more an audit list to show what keys can do.

1 Like