java.lang.IllegalStateException: Queue full

[rohithbv1991]

java.lang.IllegalStateException: Queue full
    at java.util.AbstractQueue.add(AbstractQueue.java:98)
    at com.github.brainlag.nsq.Connection.incoming(Connection.java:115)
    at com.github.brainlag.nsq.netty.NSQHandler.lambda$channelRead0$3(NSQHandler.java:41)
    at io.netty.util.concurrent.SingleThreadEventExecutor.runAllTasks(SingleThreadEventExecutor.java:358)
    at io.netty.channel.nio.NioEventLoop.run(NioEventLoop.java:357)
    at io.netty.util.concurrent.SingleThreadEventExecutor$2.run(SingleThreadEventExecutor.java:112)
    at io.netty.util.concurrent.DefaultThreadFactory$DefaultRunnableDecorator.run(DefaultThreadFactory.java:137)
    at java.lang.Thread.run(Thread.java:748)

Went through the code and found the following line in Connection.java :

private final LinkedBlockingQueue<NSQFrame> responses = new LinkedBlockingQueue<>(1);

Wanted to understand why this is initialized with capacity 1 and what could potentially be causing this error. Could it be related to high rate of polling NSQ? We deal with a lot of data.

[yyqian]

Same here.

I find this problem is caused by the 'no-waiting' command and ErrorFrame processing logic

For example:

If I call FIN command multiple times, nsqd will response "E_FIN_FAILED FIN 0a8c981c6da8c000 failed ID not in flight"

The problem is:

1. the FIN command is sent but not waiting for response
2. Read Channel received ErrorFrame, this frame is added to responses queue, but no consumer is waiting on this queue
3. if 1 and 2 happens again, the responses queue is overflow

connection.command(NSQCommand.finish(this.id))

public ChannelFuture command(final NSQCommand command) {
        return channel.writeAndFlush(command);
}

        if (frame instanceof ErrorFrame) {
            if (errorCallback != null) {
                errorCallback.error(NSQException.of((ErrorFrame) frame));
            }
            responses.add(frame);
            return;
        }

[guptaaditya13]

@rohithbv1991 I was facing the same issue again, I tracked it and read the the whole source code then came here to raise the issue and found this. :joy:

[creepyghost]

Faced the same problem. It crashes after the second timed out message.

[pranavanurag]

hello i'm here too now

[curefit-aditya-gupta]

Go to RabbitMQ, NSQ is dead.

[yuexinok]

Connection

I met too，because message.finished() Execute many times

[cssword]

This forked from nsqio/TrendrrNSQClient , and in that project,

responses.add(frame);

has been replaced by

responses.offer(frame,1,...);

Hope help anyone.