The most probable cause is the consumer process being swapped out, and unable to read messages in time, which queue up and exceed the default 1,000 message "high water mark" receive buffer.
We may be able to help if you tell us: - what version of 0MQ you are using - the operating system and hardware configurations - the message rate (messages per second) and typical message size - whether consumers may be fighting for CPU cores with other processes - precisely the types of sockets you are using - whether you're losing one in every two messages, or bursts of messages Also, please read the solver here: http://zguide.zeromq.org/page:all#Missing-Message-Problem-Solver Thanks Pieter On Mon, Jan 28, 2013 at 6:32 PM, Varun Vijayaraghavan <[email protected]> wrote: > Hi all, > > A quick overview of our setup: We have many publishers writing to a set of > fixed IP forwarders, and we have various consumer processes that subscribe > from the forwarder. > > On some of our processes, which incidentally run on smaller instances, we > see that the message count in the consumer suddenly drop to about 50%. This > happens once a week, and does not get fixed by itself till we restart the > consumer process. > > Is this expected behavior related to smaller machines, or .. something else? > Also, could someone explain the mechanism that would cause the ping count to > drop like that? > > Thanks much! > > > > -- > - varun :) > > _______________________________________________ > zeromq-dev mailing list > [email protected] > http://lists.zeromq.org/mailman/listinfo/zeromq-dev > _______________________________________________ zeromq-dev mailing list [email protected] http://lists.zeromq.org/mailman/listinfo/zeromq-dev
