Spring Boot - Limit on number of connections created

Question

I developed a microservice using Spring Boot. I was performance testing the service by stubbing the backend calls. When I looked at the thread count , I see that the maximum number of threads that created to the service is 20 at any point in time even though the number of calls made is much higher. Are there any limitations with respect to number of calls that can be made to a microservice developed using Spring Boot. Please can you guide in what steps I need to follow to troubleshoot / increase the number of connections accepted by the service?

Answer 1

This setting is derived from the embedded container (tomcat, jetty...).

Tomcat's number of threads

You may specify this property in your application.properties

server.tomcat.threads.max=400

You say you counted 20 threads, however according to this other stackoverflow question/answer, the default number of thread should be 200 with tomcat, since server.tomcat.max-threads's default value is 0. See tomcat's documentation:

The maximum number of request processing threads to be created by this Connector, which therefore determines the maximum number of simultaneous requests that can be handled. If not specified, this attribute is set to 200. If an executor is associated with this connector, this attribute is ignored as the connector will execute tasks using the executor rather than an internal thread pool.

Also, the property for:

• undertow: server.undertow.worker-threads

• jetty: server.jetty.acceptors

You'll find the list of properties in Spring's documentation

Answer 2

Maybe, you can have a look at the springboot's config

server.tomcat.accept-count=100 # Maximum queue length for incoming connection requests when all possible request processing threads are in use.
server.tomcat.additional-tld-skip-patterns= # Comma-separated list of additional patterns that match jars to ignore for TLD scanning.
server.tomcat.background-processor-delay=10s # Delay between the invocation of backgroundProcess methods. If a duration suffix is not specified, seconds will be used.
server.tomcat.basedir= # Tomcat base directory. If not specified, a temporary directory is used.
server.tomcat.max-connections=10000 # Maximum number of connections that the server accepts and processes at any given time.
server.tomcat.max-http-header-size=0 # Maximum size in bytes of the HTTP message header.
server.tomcat.max-http-post-size=2097152 # Maximum size in bytes of the HTTP post content.
server.tomcat.max-threads=200 # Maximum amount of worker threads.
server.tomcat.min-spare-threads=10 # Minimum amount of worker threads.
server.tomcat.port-header=X-Forwarded-Port # Name of the HTTP header used to override the original port value.
server.tomcat.protocol-header= # Header that holds the incoming protocol, usually named "X-Forwarded-Proto".
server.tomcat.protocol-header-https-value=https # Value of the protocol header indicating whether the incoming request uses SSL.
server.tomcat.redirect-context-root=true # Whether requests to the context root should be redirected by appending a / to the path.
server.tomcat.remote-ip-header= # Name of the HTTP header from which the remote IP is extracted. For instance, `X-FORWARDED-FOR`.
server.tomcat.resource.cache-ttl= # Time-to-live of the static resource cache.
server.tomcat.uri-encoding=UTF-8 # Character encoding to use to decode the URI.
server.tomcat.use-relative-redirects= # Whether HTTP 1.1 and later location headers generated by a call to sendRedirect will use relative or absolute redirects.

Answer 3

While the accepted answer is very useful, I recently experienced what I believe to be the same problem as the original poster. This is the only search result I could find that directly correlated with my experience, so I thought I'd add my solution in case it helps someone.

In my case, the observed concurrency limit of 20 was imposed by the default setting of 20 for themaxConcurrentStreamExecution property in org.apache.coyote.http2.Http2Protocol.

If you're experiencing this problem and you're using HTTP/2, there's a good chance that increasing maxConcurrentStreamExecution will help.

You can find more info in the Tomcat Configuration Reference, which actually states that this should be set to 200 by default (not 20). You can definitely see the default setting of 20 in org.apache.coyote.http2.Http2Protocol, though, so I'm not sure if this is a typo or just something that presents itself differently in the embedded version of Tomcat.

Answer 4

If you have actuator you can see the metrics

/actuator/metrics/tomcat.threads.config.max

{
  "name": "tomcat.threads.config.max",
  "description": null,
  "baseUnit": null,
  "measurements": [{
    "statistic": "VALUE",
    "value": 200.0
  }],
  "availableTags": [{
    "tag": "name",
    "values": ["http-nio-8080"]
  }]
}

The actual value tomcat decided to create? /actuator/metrics/tomcat.threads.current

you may see 10 there depending on the load

spring boot never seems to fully use max-threads you can however start with more

server:
   tomcat:
     min-spare-threads: 40

Answer 5

Increase maxConcurrentStreamExecution (set 200) for HTTP/2 in Spring Boot 2:

@Bean
public WebServerFactoryCustomizer<TomcatServletWebServerFactory> containerCustomizer() {
    return new WebServerFactoryCustomizer<TomcatServletWebServerFactory>() {
        @Override
        public void customize(TomcatServletWebServerFactory factory) {
            factory.addConnectorCustomizers(new TomcatConnectorCustomizer() {
                @Override
                public void customize(Connector connector) {
                    Arrays.stream(connector.getProtocolHandler().findUpgradeProtocols())
                        .filter(upgradeProtocol -> upgradeProtocol instanceof Http2Protocol)
                        .map(upgradeProtocol -> (Http2Protocol) upgradeProtocol)
                        .forEach(http2Protocol -> http2Protocol.setMaxConcurrentStreamExecution(200));
                }
            });
        }
    };
}