Set up Microsoft Teams notifications using the discourse-chat-integration plugin

This #howto describes how to set up the Micorosoft Teams provider included in discourse-chat-integration .

Install

If you haven’t already, install the discourse-chat-integration plugin

Setup Microsoft Teams

  1. Navigate to the channel where you want to add the webhook and select (•••) More Options from the top navigation bar.
  2. Choose Connectors from the drop-down menu and search for Incoming Webhook .
  3. Select the Configure button, provide a name, and, optionally, upload an image avatar for your webhook.
  4. The dialog window will present a unique URL that will map to the channel. Make sure that you copy and save the URL —you will need to provide it to the outside service.
  5. Select the Done button. The webhook will be available in the team channel.

Setup Discourse

  1. In your site settings, enable the chat integration teams enabled setting
  2. Go to the Plugins tab, then Chat Integrations
  3. Go to the Microsoft Teams tab, and click Add Channel
  4. Enter the name of your teams channel (e.g. discourse)
  5. Paste the Webhook URL that you copied earlier, then click Save Channel
  6. Use the Test button on the channel you just created to check that everything’s working. You should see something like this in the chatbox of your microsoft teams channel

  1. Setup some Rules for your channel, according to the discourse-chat-integration 23 instructions.
  2. Celebrate :tada:
11 Likes

Just tried it out and it seemed to work at first, but only the first time I clicked on the test button.
Rules are set up as follows:

  • Type: normal
  • Filter: All topics(?) and replies (translating from German here)
  • Category: Sandbox

When I wrote a reply to the test topic in my Sandbox category I got a warning sign when I went to my chat integrations, but when clicking on the warning triangle it says unknown error... with the error message bein null.

When I try to send another test notification it gives me another error 500 internal server error

Any hints on where to start trouble shooting this are welcome. If this should be posted elsewhere (e.g. under the chat integration plugin page), or made a separate topic, feel free to move it accordingly or let me know.

1 Like

Hi @SHilser - please can you check /logs on your site, and see if there is anything relevant there?

1 Like

There’s two things that seem relevant.

  • Socket Error
  • Internal Server Error

I copied the logs below (I replaced my domain name with mydomain.tld)

Socket Errror

Message (4 copies reported)

SocketError (Failed to open TCP connection to outlook.office.com:443 (getaddrinfo: Temporary failure in name resolution))
app/controllers/application_controller.rb:340:in `block in with_resolved_locale'
app/controllers/application_controller.rb:340:in `with_resolved_locale'
lib/middleware/omniauth_bypass_middleware.rb:68:in `call'
lib/content_security_policy/middleware.rb:12:in `call'
lib/middleware/anonymous_cache.rb:336:in `call'
config/initializers/100-quiet_logger.rb:19:in `call'
config/initializers/100-silence_logger.rb:31:in `call'
lib/middleware/enforce_hostname.rb:22:in `call'
lib/middleware/request_tracker.rb:176:in `call'

Backtrace

/usr/local/lib/ruby/2.6.0/net/http.rb:949:in `rescue in block in connect'
/usr/local/lib/ruby/2.6.0/net/http.rb:946:in `block in connect'
/usr/local/lib/ruby/2.6.0/timeout.rb:93:in `block in timeout'
/usr/local/lib/ruby/2.6.0/timeout.rb:103:in `timeout'
/usr/local/lib/ruby/2.6.0/net/http.rb:945:in `connect'
/usr/local/lib/ruby/2.6.0/net/http.rb:930:in `do_start'
/usr/local/lib/ruby/2.6.0/net/http.rb:919:in `start'
/usr/local/lib/ruby/2.6.0/net/http.rb:1470:in `request'
rack-mini-profiler (2.0.4) lib/patches/net_patches.rb:19:in `block in request_with_mini_profiler'
rack-mini-profiler (2.0.4) lib/mini_profiler/profiling_methods.rb:33:in `step'

Env

HTTP HOSTS: mysite.tld

Internal Server Error

Message (4 copies reported)

Error: Internal Server Error
Url: https://mysite.tld/assets/ember_jquery-1ed3f3559e6f967733b4088aa729ff7039dff2c09c5a5f787a214b016f58aabc.js
Line: 1
Column: 267890
Window Location: https://mysite.tld/admin/plugins/chat/teams

Backtrace

o/t</<@https://mysite.tld/assets/ember_jquery-1ed3f3559e6f967733b4088aa729ff7039dff2c09c5a5f787a214b016f58aabc.js:1:267890
o/t<@https://mysite.tld/assets/ember_jquery-1ed3f3559e6f967733b4088aa729ff7039dff2c09c5a5f787a214b016f58aabc.js:1:267993
o@https://mysite.tld/assets/ember_jquery-1ed3f3559e6f967733b4088aa729ff7039dff2c09c5a5f787a214b016f58aabc.js:1:268074
trigger@https://mysite.tld/assets/ember_jquery-1ed3f3559e6f967733b4088aa729ff7039dff2c09c5a5f787a214b016f58aabc.js:1:545343
A</e._onError/<@https://mysite.tld/assets/ember_jquery-1ed3f3559e6f967733b4088aa729ff7039dff2c09c5a5f787a214b016f58aabc.js:1:550189
f</t.invoke@https://mysite.tld/assets/ember_jquery-1ed3f3559e6f967733b4088aa729ff7039dff2c09c5a5f787a214b016f58aabc.js:1:490190
f</t.flush@https://mysite.tld/assets/ember_jquery-1ed3f3559e6f967733b4088aa729ff7039dff2c09c5a5f787a214b016f58aabc.js:1:489198
p</t.flush@https://mysite.tld/assets/ember_jquery-1ed3f3559e6f967733b4088aa729ff7039dff2c09c5a5f787a214b016f58aabc.js:1:491205
t</t._end@https://mysite.tld/assets/ember_jquery-1ed3f3559e6f967733b4088aa729ff7039dff2c09c5a5f787a214b016f58aabc.js:1:496596Here
t</t.end@https://mysite.tld/assets/ember_jquery-1ed3f3559e6f967733b4088aa729ff7039dff2c09c5a5f787a214b016f58aabc.js:1:493157

Env

HTTP HOSTS: mysite.tld

Thanks for the help :slight_smile:

1 Like

That suggests that your server is having trouble with DNS resolution. Are you using a standard docker-based installation? Or is this in development?

2 Likes

Just set it up about a week ago on a separate DigitalOcean droplet (one-click installation, now I’m on Version: 2.6.0.beta1 (310952fd6a). I use a subdomain and have the DNS records also on DigitalOcean. I have an A-Record, pointing to the IP, MX-Record pointing to the sub-domain name for reply by email as per these instruction). Only thing there is that I don’t have a DMARC Record published.

That’s all I can think of that I did in relation to DNS stuff. However, so far everything worked as expected. The site is set up, I can reach it via the sub-domain name, log-in, create topics (even via email), etc.

I’m just startled the first message made it through, but subsequent ones didn’t :man_shrugging:

1 Like

When I say DNS resolution, I mean your server is having trouble contacting outlook.office.com. The setup of your forum’s DNS records shouldn’t affect that.

Can you try connecting to your server via SSH and running

host outlook.office.com

You should get something like

outlook.office.com is an alias for substrate.office.com.
substrate.office.com is an alias for substrate.ms-acdc.office.com.
substrate.ms-acdc.office.com is an alias for afd-k.office.com.
afd-k.office.com is an alias for outlook-office-com.k-0002.k-msedge.net.
outlook-office-com.k-0002.k-msedge.net is an alias for k-0002.k-msedge.net.
k-0002.k-msedge.net has address 13.107.18.11
k-0002.k-msedge.net has IPv6 address 2620:1ec:c::11
2 Likes

When I enter

host -d outlook.office.com

I get the following:

Trying "outlook.office.com"
Host outlook.office.com not found: 2(SERVFAIL)
Received 36 bytes from 127.0.0.53#53 in 43 ms

When adding port :443 I get another error

Trying "outlook.office.com:443"
Host outlook.office.com:443 not found: 3(NXDOMAIN)
Received 40 bytes from 127.0.0.53#53 in 3 ms
Received 40 bytes from 127.0.0.53#53 in 3 ms

For other domains like google.com or the domain of my discourse instance in question (h-da.transformative-praxis.cc), or other sites I get the expected output similar to what you describe.

I did some search for the first error, but I’m still not sure about its source. I’ve read the following, but wasn’t able to judge, if this is actually relevant to my problem:

I kept retrying it and this time I got the results you described when typing

host outlook.office.com

So everything seemed fine at this point. I’ve repeated that many times in a row to make sure (like around 15 times with an interval of around 10 seconds) and at some point I got the same servfail error as before. But mostly it worked.

The opposite is the case when I try to send a test message. It maybe works one out of ten times. The other times I get the same errors as before. And when I post a regular message that fits to the rule I’ve set for the channel, I don’t get a notification, nor an error message in the logs.
It just seems so random to me that I can’t find out what causes it.

I kept suspecting my setup with two droplets for each subdomain and two respective A-Records, plus one set of NS-records for my main domain pointing to ns1.digitalocean.com. So I tried out dnsstuff for my site getting a Fail for the SOA record check and whatsmydns.net for my NS records, but I still couldn’t figure out any possible connections to my problem.

At this point I’m totally confused who’s causing the error, me having made some wrong DNS settings (which shouldn’t affect it according to @david), server settings/issues, or is it just MS-Teams (outlook.office.com) acting up, or blocking me in some way?

Any hints for what I could check/do to find the error are appreciated, including where else to go and ask, if you think this is beyond what you can do to help me here.

Thanks a lot for all the help so far.