I have a Linux server running Nginx and from time to time it starts throwing a 502 error when trying to access it throw HTTP, and after a while gets back to normal.
I've checked the different logs (Nginx, PHP and MySQL) and didn't find anything that could help me find what's causing this.
Any idea where I should look or if it's possible to set any logging/monitorinh tool to help me out?
Thanks,
502 is a Bad Gateway error. You may want to use a log monitoring tool to see these types of errors. You may also want to send the PHP errors to see if they are generating fatal errors at the same time. I personally use Loggly for this purpose, but you can use other tools available in the market.
Also, I would recommend logging Nginx as JSON, and adding to the normal metrics, the uptime for the upstream servers, so that you can see if the 502 errors are also correlated with deteriorating performance.
If you haven't read this already, I recommend starting here:
https://www.nginx.com/resources/admin-guide/logging-and-monitoring/
Related
I am running Ghost in development mode on FastComet hosting company. I dont know why but suddenly when i try to access my website I recieve the error
Service Unavailable
The server is temporarily unable to service your request due to maintenance downtime or capacity problems. Please try again later.
Additionally, a 503 Service Unavailable error was encountered while trying to use an ErrorDocument to handle the request.
If I do ghost restart it says that ghost is turned off, so it’ll turn on and everything comes back to normal. But after some minutes the 503 error comes back.
Does anyone now how can I find out the cause?
Not sure if this is a Docusign or ngix question.I'm working on integrating an application with Docusign and I keep seeing this somewhat vague error below in the Docusign Connect logs. In our nginx logs I see that a POST to our application's /webhook endpoint was attempted but doesn't go through. I've specified TLS 1.2 and have tried increasing our nginx timeout but that doesn't seem to fix it.
One theory I have is that our server's certificate isn't chained to a Microsoft trusted CA but I would expect a different error if that was the case.
Any help or guidance would be greatly appreciated.
This most likely would require you to get the IT folks managing this server and networking involved. And yes, they may need to install a certificate, but other errors can be related to the firewall blocking certain requests, an anti-virus blocking requests and even DNS related error preventing the HTTP request from being sent and received by the server.
We highly recommend to use a public cloud for Connect and we have plenty of examples how this can work, while still having your code run on your own IT server.
I have done jmeter load testing and it was working fine.
Now i am doing jmeter load testing with ngnix by configuring two server details .
I am testing my application by giving the 1000 concurrent users with nginx url.
Sometimes it was working and sometimes not.
Why it is behaving like that??
Check the logs — nginx logs, your app and server logs and JMeter logs. Probably you can find answer there:
Check nginx.conf and find error_log settings to get current nginx error log location.
What do you mean "sometimes not"? What kind of errors, statuses and responces do you get?
You may post in here.
I configured my gitlab instance with an external nginx server according this documentation and following nginx configuration.
All is working fine, but when i try to change some profile settings in the UI I get an connection reset.
The problem only occurs on multipart form POST/PUT operations.
The funny thing is, that there are no log entries in any gitlab log files, so i think there is a problem in the nginx configuration somewhere.
Please when someone has any idea on how to further analyse the problem.
Thanks
The problem was the MTU setting for the TCP-IP Stack on my windows machine. I switched from 1500 to 1300 and now all is working fine again.
On our openshift (jboss) application are 16 gears available.
But openshit does not scale.
If 50 users are online on our application (we can track it with google analytics), we are getting the following error:
503 errors (Service Temporarily Unavailable)
Your question does not give enough information about what does not work. Have you looked at the logs to see if you are getting error. You can use rhc tail command to look at the logs. Have you done some analysis on what is the bottleneck of your application. In OpenShift database does not scale. How many database connections are you creating? Have you tried configuring number of JBoss instance? How are you load testing the application? What behavior are you seeing?