htrace-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Colin McCabe <cmcc...@apache.org>
Subject Re: Apache HTrace question
Date Sat, 25 Jun 2016 20:14:39 GMT
Hi Nuc,

Welcome!  If you would like to try out HTrace on a real system, Hadoop
is the best choice.

There is some information about setting up HTrace+Hadoop here:
http://blog.cloudera.com/blog/2015/12/new-in-cloudera-labs-apache-htrace-incubating/

If you would like to take a look at the code for HTrace in Hadoop, a
good starting point is here:
https://github.com/apache/hadoop/commit/892ade689f9bcce76daae8f66fc00a49bee8548e

Perhaps one point that might be confusing is that in general, HTrace is
agnostic to the choice of RPC systems.  Also, when HTrace is integrated
into an RPC system, the source code for that integration will be in the
repository where the RPC system is implemented, not in the HTrace code
repository.  In general the integration with RPC is pretty simple-- just
a matter of adding an optional field with an extra 128 bits for a parent
span ID.

best,
Colin


On Sat, Jun 25, 2016, at 04:23, Nuc Andrei wrote:
> Hi,
> 
> I am working on a thesis called “Tracing in distributed systems” and saw
> your project.
> 
> I have a short question. Is there any code example with multiple nodes
> involved in the tracing flow? Based on tests found for htrace-core
> component, I did not manage to build an example that could work in a
> distributed environment.
> 
> Thank you.
> 
> Regards
> Andrei Nuc,
> Technical University of Cluj-Napoca
> Romania.
> 

Mime
View raw message