I recreated this exception with a very simple test-case. I took the
consumer code pasted earlier in the thread and just added a producer
sending a large message. I lowered the minLargeMessageSize to make it
faster. I thought I still had that code laying around somewhere, but I
can't find it at the moment.
Justin
On Sun, Feb 20, 2022 at 9:03 AM Clebert Suconic <clebert.suco...@gmail.com>
wrote:
> I have seen (and fixed) cases where the large message file is gone. I
> would need a reproducer creating the issue from scratch (send and consume)
>
> Typically it could be associated with paging ? Did you have the
> destination in page mode ?
>
> On Thu, Feb 17, 2022 at 6:53 PM Tim Jones <t...@abcwxy.com> wrote:
>
> > The client code below the stack trace - will reproduce it - the
> > msg.getDataBuffer() inside the handler of that code will trigger it when
> a
> > large message is sent to the address. The exception in question
> > (IndexOutOfBoundsException) is being caught in line 249 of the apache
> > CoreMessage code... and turned into a logged warning on line 250. The
> > CoreMessage code then returns the buffer from getReadOnlyBuffer() (which
> > appears to be fine from a quick survey - data also seems to be preserved
> > and accessible from client code) - it is just not clear what the intent
> is
> > if the exception code is executed... is it a "don't worry about it" - or
> a
> > "something is wrong here" and I should concern myself with something?
> (the
> > warning level is making me believe that it is more than a "don't worry
> > about it" - but the fact that the exception was caught and a valid buffer
> > is returned makes me think it is just a fallback for the other choices of
> > buffers - and I should not worry about it?).
> >
> > Thanks for any insight you may have....
> >
> > logged warning from the caught exception is:
> >
> > 16:26:50.923 [Thread-0 (ActiveMQ-client-global-threads)] WARN
> > org.apache.activemq.artemis.core.message.impl.CoreMessage -
> readerIndex(4)
> > + length(270740) exceeds writerIndex(4):
> > UnpooledByteBufAllocator$InstrumentedUnpooledUnsafeHeapByteBuf(ridx: 4,
> > widx: 4, cap: 270740)
> > java.lang.IndexOutOfBoundsException: readerIndex(4) + length(270740)
> > exceeds writerIndex(4):
> > UnpooledByteBufAllocator$InstrumentedUnpooledUnsafeHeapByteBuf(ridx: 4,
> > widx: 4, cap: 270740)
> > at
> >
> >
> io.netty.buffer.AbstractByteBuf.checkReadableBytes0(AbstractByteBuf.java:1442)
> > at
> >
> >
> io.netty.buffer.AbstractByteBuf.checkReadableBytes(AbstractByteBuf.java:1428)
> > at io.netty.buffer.AbstractByteBuf.readBytes(AbstractByteBuf.java:937)
> > at
> >
> >
> org.apache.activemq.artemis.core.client.impl.ClientMessageImpl$DecodingContext.readInto(ClientMessageImpl.java:407)
> > at
> >
> >
> org.apache.activemq.artemis.core.message.impl.CoreMessage.getLargeMessageBuffer(CoreMessage.java:264)
> > at
> >
> >
> org.apache.activemq.artemis.core.message.impl.CoreMessage.getDataBuffer(CoreMessage.java:241)
> > at io.m45.sart.Main.lambda$main$0(Main.java:57)
> > at
> >
> >
> org.apache.activemq.artemis.core.client.impl.ClientConsumerImpl.callOnMessage(ClientConsumerImpl.java:1013)
> > at
> >
> >
> org.apache.activemq.artemis.core.client.impl.ClientConsumerImpl$Runner.run(ClientConsumerImpl.java:1133)
> > at
> >
> >
> org.apache.activemq.artemis.utils.actors.OrderedExecutor.doTask(OrderedExecutor.java:42)
> > at
> >
> >
> org.apache.activemq.artemis.utils.actors.OrderedExecutor.doTask(OrderedExecutor.java:31)
> > at
> >
> >
> org.apache.activemq.artemis.utils.actors.ProcessorBase.executePendingTasks(ProcessorBase.java:65)
> > at
> >
> >
> java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1128)
> > at
> >
> >
> java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:628)
> > at
> >
> >
> org.apache.activemq.artemis.utils.ActiveMQThreadFactory$1.run(ActiveMQThreadFactory.java:118)
> > 16:26:50.924 [Thread-0 (ActiveMQ-client-netty-threads)] DEBUG
> > org.apache.activemq.artemis.core.client.impl.ClientConsumerImpl - Sending
> > 65962 from flow-control
> >
> > Simple Client Code below:
> >
> >
> > public class Main {
> >
> > private final static String ACCEPTOR = "tcp://localhost:9322";
> > private final static String QUEUE="service.images.dev::
> > service.images.dev";
> >
> > public static void main(String[] args) throws Exception {
> >
> > final String username = null;
> > final String password = null;
> >
> > var serverLocator = ActiveMQClient
> > .createServerLocator(ACCEPTOR)
> > .setBlockOnDurableSend(true)
> > .setBlockOnNonDurableSend(true);
> >
> > final var sessionFactory = serverLocator.createSessionFactory();
> >
> > final var xa = false;
> > final var autoCommitSends = true;
> > final var autoCommitAcks = true;
> > final var ackBatchSize = serverLocator.getAckBatchSize();
> > final var preAcknowledge = serverLocator.isPreAcknowledge();
> > final var clientSession = sessionFactory.createSession(
> > username,
> > password,
> > xa,
> > autoCommitSends,
> > autoCommitAcks,
> > preAcknowledge,
> > ackBatchSize
> > );
> >
> > var queueQueryResult =
> > clientSession.queueQuery(SimpleString.toSimpleString(QUEUE));
> > if (!queueQueryResult.isExists()) {
> > clientSession.createQueue(_ServiceQueueConfiguration(new
> > SimpleString(QUEUE)));
> > }
> >
> > final var consumer = clientSession.createConsumer(QUEUE);
> >
> > clientSession.start();
> >
> > consumer.setMessageHandler((msg) -> {
> >
> > System.out.println("Received: "+msg.getBodySize());
> > msg.getDataBuffer();
> >
> > });
> >
> > while(true) {
> > Thread.sleep(1000);
> > }
> >
> > }
> >
> > private static QueueConfiguration
> > _ServiceQueueConfiguration(SimpleString queueName) {
> > final var config = new QueueConfiguration(queueName);
> > config.setMaxConsumers(1);
> > config.setPurgeOnNoConsumers(false);
> > config.setDurable(false);
> > config.setAutoDelete(false);
> > config.setRoutingType(RoutingType.MULTICAST);
> > return config;
> > }
> >
> >
> > On Thu, Feb 17, 2022 at 2:28 PM Justin Bertram <jbert...@apache.org>
> > wrote:
> >
> > > Typically an IndexOutOfBoundsException indicates a bug. Do you have a
> way
> > > to reproduce this?
> > >
> > >
> > > Justin
> > >
> > > On Thu, Feb 17, 2022 at 3:17 PM Tim Jones <t...@abcwxy.com> wrote:
> > >
> > > > This seems to appear on larger messages only - I am getting a warning
> > > when
> > > > calling getDataBuffer (2.20.0 Artemis Client). Curious if there is
> > > > something I may be missing - or if this is completely ignorable?
> > Thanks -
> > > > Tim
> > > >
> > > > 2022-02-17T21:06:11,908+01:00 [Thread-2
> > (ActiveMQ-client-global-threads)]
> > > > WARN o.a.a.a.c.m.i.CoreMessage [Thread-2
> > > (ActiveMQ-client-global-threads)]
> > > > readerIndex(270740) + length(270740) exceeds writerIndex(271572):
> > > > UnpooledByteBufAllocator$InstrumentedUnpooledUnsafeHeapByteBuf(ridx:
> > > > 270740, widx: 271572, cap: 271572)
> > > > java.lang.IndexOutOfBoundsException: readerIndex(270740) +
> > length(270740)
> > > > exceeds writerIndex(271572):
> > > > UnpooledByteBufAllocator$InstrumentedUnpooledUnsafeHeapByteBuf(ridx:
> > > > 270740, widx: 271572, cap: 271572)
> > > > at
> > > >
> > > >
> > >
> >
> io.netty.buffer.AbstractByteBuf.checkReadableBytes0(AbstractByteBuf.java:1442)
> > > > at
> > > >
> > > >
> > >
> >
> io.netty.buffer.AbstractByteBuf.checkReadableBytes(AbstractByteBuf.java:1428)
> > > > at
> io.netty.buffer.AbstractByteBuf.readBytes(AbstractByteBuf.java:937)
> > > > at
> > > >
> > > >
> > >
> >
> org.apache.activemq.artemis.core.client.impl.ClientMessageImpl$DecodingContext.readInto(ClientMessageImpl.java:407)
> > > > at
> > > >
> > > >
> > >
> >
> org.apache.activemq.artemis.core.message.impl.CoreMessage.getLargeMessageBuffer(CoreMessage.java:264)
> > > > at
> > > >
> > > >
> > >
> >
> org.apache.activemq.artemis.core.message.impl.CoreMessage.getDataBuffer(CoreMessage.java:241)
> > > >
> > >
> >
> --
> Clebert Suconic
>
