The fail method of the spout gets called on message timeouts. It will not stop/suspend processing of the current message flow. Everything happens asynchronously, read about ackers in storm docs.
Thanks and Regards, Devang On 10 Jun 2015 08:45, "Ganesh Chandrasekaran" <[email protected]> wrote: > I am also seeing the same behavior on 0.8.2 version. I am not sure if I > am understanding the purpose of message timeouts correctly. > > > > Thanks, > > Ganesh > > > > *From:* Ganesh Chandrasekaran [mailto:[email protected]] > *Sent:* Tuesday, June 09, 2015 5:01 PM > *To:* [email protected] > *Subject:* Multilang: message timeout and tick tuple question > > > > Hi all, > > > > I was reading the attached email thread that was discussed here before and > I have few questions about message timeouts and tick tuples. > > > > I am doing some testing using 0.9.4 version and I have my > topology.message.timeout.secs = 60 > > > > I have a bolt that is written using multilang protocol in PHP and is doing > the following: > > 1. Log a message to indicate process started > > 2. Sleep for 150 seconds > > 3. Log a message to indicated process ended > > > > I would have expected the message to expire after 60 seconds but I see the > process completing successfully by logging the 3rd step as well. > > > > My questions – > > 1. What exactly does the message timeout do? Because it looks like > it is doing nothing in terms of timing out the message. Also, I have > *“topology.enable.message.timeouts” > set to true* > > 2. Do I need to call something in code to explicitly invoke the a > function that would eventually do the check to expire the message? > > 3. What does the tick tuple do? Initially I had not set > *“topology.tick.tuple.freq.secs”* at all. So from reading the attached > email conversation I think it was set to message timeout seconds by > default. Later I set it to 2 secs and my worker won’t start and keeps > failing. What is an expected value for this and when should we be > configuring this? > > > > P.S: I read documentations but they don’t seem to give me a detailed > understanding > > > > Thanks, > > Ganesh >
