Problem with Nginx - Would it be multi_accept? only accepts 1 requisition at a time
I have 1 system with 2 instances, 1 clouds for the DB and another cloud for the system and interface.
It turns out that when I am going to execute a task, for example export 100 products … The whole system freezes and no one else can access my site .. OBS: the current task continues to run and my system access only comes back to everyone when it completes action. this happens on all long tasks … like importing, exporting and connecting clients.
At that time I was exporting some products, so I could see the PUT screen connected to my cloud
While I'm exporting, everyone trying to access the system can not get it .. and the message below is displayed in the chrome … Waiting for available socket.
I do not know what to do, I do not understand much about Ngixn, I'll show you how my Ngixn configuration file is.
If anyone can help me, I'll be grateful! I tried to make several changes and nothing solves … I do not know what to do … everything I'm going to do I have to wait to give 1 of tomorrow not to gain access to any client.
The cloud of the DB has 4 cores, and the cloud of the system/interface has 2 cores~~~~
I want to preface this by saying that I've not personally encountered this situation before, and what I'm giving you below is the result of online research. Hopefully it'll be a good place to start and will get you pointed in the right direction.
I'm also going to add some additional tags to this post so that it gets more visibility to our Community - that way it'll be more readily available to users who may have run into this issue before.
One of the first things I noticed is that the "Waiting for available socket" error is happening on Chrome - is that correct? I found this that goes into it a bit:
Another page I found on Stack Exchange gives some other ideas on how to look into it, with some detailed potential fixes:
I hope this helps!