Vladi, check the limit on open files in Kafka and supervisor servers. also check if there is any IO overhead.
The processing may completely stop if there are too many files waiting to be written. run the top command -Manoj On Tue, Oct 21, 2014 at 2:08 PM, Vladi Feigin <[email protected]> wrote: > One of two workers doesn't read the data from kafka at all. In this worker > all succeeding bolts show 0 in emitted / transferred in UI > This happens after a few hours of successful running. We don't use acks in > this topology > Vladi > > On Tue, Oct 21, 2014 at 11:55 PM, saiprasad mishra < > [email protected]> wrote: > >> Is the topology not reading from kafka at all and not marking the offset >> at all. >> Recently I ran into similar issue which was because of buggy code in my >> toology, where the ack was happening 2 times instead of once, one in the >> tey block and one in the finally block inside the execute method. This was >> causing the messages in pending state and spout was not emitting as the >> pending messages were not completed. Wondering if you have similar issues >> in your code >> >> On Tue, Oct 21, 2014 at 1:50 PM, Vladi Feigin <[email protected]> wrote: >> >>> H All, >>> >>> We're experiencing very strange topology behavior when after a few days, >>> sometimes hours (looks like during a peak load) the spouts get stuck. >>> The data stops streaming and we lose a lot of data. >>> We read the data from kafka (use kafka spout). Storm version 0.8.2 >>> Does someone have something similar? What can be a problem? >>> >>> Appreciate any help you can provide! >>> Thank you in advance >>> Vladi >>> >>> >> >
