longpoll | PubSub queuing with long-polling subscribers | Pub Sub library

I have created a signalr selfhosted webapp service that connects to on prem signalr service using hybrid connection. This self hosted service then broadcast the changes to front end and back to on prem signalR. The self hosted service is fronted by a front door.

Everything works fine locally with long polling (front door doesn't support websockets), but doesn't work when deployed in Azure. Error: Failed to start the transport 'LongPolling': Error: No Connection with that ID: Status code '404'

Anything i am missing?

Issue: Socket rejoins again and again which causes Live View to emit events itself periodically.

Scenario: I have index page which contains phx-change events. When I left the page idle for sometime, the events started to triggered automatically periodically. It is not limited to single page but it happen on each live view page, I googled the issue but couldn’t find any solution.

• 1 reason I noticed that sometimes heartbeat stops for more than 1 minute(timeout limit is 1 minute), in this case socket rejoined again.

• I increased timeout at client side as well at server side in Endpoint to check if this is the only issue but it didn’t work and socket behaved the same.

• Did anyone face same issue before, what could be a possible reasons and how to avoid this issue, any suggestion?

Stack: PETAL

Elixir: 1.11 Erlang: 23.0 phoenix: 1.5.3 phoenix_live_view: 0.15.7 alpinejs “^2.8.2” Browsers: chrome, safari

I think issue is with Alpine Js but couldn’t figure out any solution.

Here is my app.js code

I'm using middleware in Redux to connect to my signalR hub, which works perfectly and I can receive messages no problem.

Now I have got to the point where I want to invoke a method on the server to "Join a Group".

What is the best way to do so? I have a connection, so recreating the connection on the action "JOIN_GROUP" seems counter intuitive.

I thought I might be onto something when I read xxx, so I could chain

when I use the WebSockets I Can't get the userId on the server

Client code:

I use this library tucnak/telebot to build a telegram bot.

Method b.Handle() have two parameters such as Handle(endpoint interface{}, handler interface{})`.

Here is the code i use for a starter

I use SignalR for my ASP.Net Application with longPolling protocol. I see that the client send the \signalr\pool request to server with the messageId form data: d-B1017AFE-SC,B8|bW,0|bs,1|bx,2.

I'm looking for the documents which describe the d-B1017AFE-SC,B8|bW,0|bs,1|bx,2 format but not yet. Please help me to understand it. Below image was captured from Chrome Dev Toll which explains what I said.

My customer uses Azure WAF to protect the site. The Azure WAF report show that above request belong to the Command Injection risk because the messageId contain |bw,|bs, |bx that match the |ps rule in Azure WAF. So I need to understand the messageId content to explain for my customer and ask them customize the rule to ignore the request.

Below is the WAF description in the report:

[{'riskGroup':'','match':'Vector Score: 5, Group Threshold: 4, Triggered Rules: 950006, Mitigated Rules: , Last Matched Message: System Command Injection','threshold':'4','type':'riskscore','version':'1','score':0,'atomics':[{'score':5,'riskGroup':'CMD-INJECTION-ANOMALY','match':'|ps','ruleName':'System Command Injection','selector':'ARGS:messageId','ruleId':'950006','version':'7'}],'actionId':'alert','ruleName':'CMD-INJECTION-ANOMALY','action':'Alert','selector':'','ruleId':'CMD-INJECTION-ANOMALY','category':'Command Injection'}]

I have been running a NestJS API in a Google Cloud App Engine Service for about 6 months now on the flex environment. I'm testing in one of my lower environments on a service hosted on the standard app engine service environment and noticed today that when I establish a websocket connection from an angular front end, it only works with the longpolling transport. In my flex environment, the websockets transport works fine.

I'm not seeing any errors on my API logging so I'm guessing the standard environment on GCP has some sort of mechanism in place that's preventing the wss traffic to connect on websockets.

It's also possible that my API just isn't logging out exceptions being thrown to the GCP logger for that event.

In the event it is something with the API is there some kind of configuration change I need to make on the API project to allow this? Otherwise, is there something I need to change when deploying to a standard environment in GCP to allow this to work?

TIA

I have 1 server, 1 Asp.Net Core app, nginx, 2 angular apps being hosting on the same Linux Debian server. My use of SignalR is as: 2 apps getting data from same SignalR hubs. I have 2 hubs. 1 is working perfectly for 2 apps, 1 is working perfectly for one app and doesn't work for another one(but they are using from practically the same code). May I ask for some ideas why this error pops up? Thank You for any help.

P.S. everything works perfectly well on local machine.

I get regularly following error message while running odoo v14 locally in docker: