Hi,

yes, it is correct to use Spout.ack() to throttle the ingestion rate.
However, you must also use Spout.fail(), in case tuples time-out or are
failed manually within a bolt. Maybe, that is the reason you are
"missing acks".

(I personally doubt, that Storm looses any tuples or ack/fail messages.)


-Matthias


On 07/28/2015 03:56 PM, Aditya Rajan wrote:
> Hello 
> 
> We are using the ack method in our spout to keep a track of unpacked
> tuples. This is so we can throttle the number of tuples our spout is
> sending. However we happen to be either losing tuples or acks as we
> notice that there is a build up in our UnAck counts.
> 
> Is it correct to use the ack method to help throttle? This never used to
> happen at lower throughputs. Could this be due to a sudden increase in
> throughputs. Also all the ackers are well below capacity. Is there any
> buffer size that we need to increase? 
> 
> Also is it possible that we are losing messages and not just acks
> 
> Thanks and Regards
> Aditya Rajan

Attachment: signature.asc
Description: OpenPGP digital signature

Reply via email to