avro-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Philip Zeyliger" <phi...@cloudera.com>
Subject Re: Review Request: AVRO-595.patch.v2.txt
Date Tue, 03 Aug 2010 06:05:30 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://review.cloudera.org/r/390/#review627
-----------------------------------------------------------


Almost there.  I tried running the tests and ran into some issues:

Testcase: testRecursingTrace took 0.006 sec
        Caused an ERROR
Connection refused
java.net.ConnectException: Connection refused


Does "ant test clean" work for you?  I think it's trying to re-use a port that it hasn't closed.



lang/java/src/java/org/apache/avro/ipc/trace/Util.java
<http://review.cloudera.org/r/390/#comment2366>

    If you can avoid this, I would avoid making this public.



lang/java/src/java/org/apache/avro/ipc/trace/Util.java
<http://review.cloudera.org/r/390/#comment2365>

    This doesn't feel public to me


- Philip


On 2010-07-30 13:11:31, Patrick Wendell wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://review.cloudera.org/r/390/
> -----------------------------------------------------------
> 
> (Updated 2010-07-30 13:11:31)
> 
> 
> Review request for Avro.
> 
> 
> Summary
> -------
> 
> Here is a patch update based on the preliminary feedback. I dropped
> the querying stuff since it's not yet used. This patch adds the logic
> behind aggregating partial spans (which will have been collected from
> several remote hosts) and creating traces (extracting RPC call trees
> from a set of spans).
> 
> We talked off-line about redundancy in the data that is collected. We
> could get away with only recording a spanID and a parentSpanID for
> each span, then we can back-out each trace by following these data (we
> get to the root of a trace when we see a null parentSpanID). By adding
> a traceID unique to each call tree, it is easier to partition the work
> of collecting spans into traces as we can easily group spans by trace
> before we try back-out a tree (perhaps this is why Dapper paper
> describes this method). It also lowers the chance of ID collision
> since spanID and parentSpanID are specific to one traceID.
> 
> 
> Diffs
> -----
> 
>   lang/java/ivy.xml a781557 
>   lang/java/src/java/org/apache/avro/ipc/trace/InMemorySpanStorage.java PRE-CREATION

>   lang/java/src/java/org/apache/avro/ipc/trace/SpanAggregator.java PRE-CREATION 
>   lang/java/src/java/org/apache/avro/ipc/trace/SpanStorage.java PRE-CREATION 
>   lang/java/src/java/org/apache/avro/ipc/trace/Trace.java PRE-CREATION 
>   lang/java/src/java/org/apache/avro/ipc/trace/TraceNode.java PRE-CREATION 
>   lang/java/src/java/org/apache/avro/ipc/trace/TracePlugin.java PRE-CREATION 
>   lang/java/src/java/org/apache/avro/ipc/trace/TracePluginConfiguration.java PRE-CREATION

>   lang/java/src/java/org/apache/avro/ipc/trace/Util.java PRE-CREATION 
>   lang/java/src/test/java/org/apache/avro/ipc/trace/TestBasicTracing.java PRE-CREATION

>   lang/java/src/test/java/org/apache/avro/ipc/trace/TestEndToEndTracing.java PRE-CREATION

>   lang/java/src/test/java/org/apache/avro/ipc/trace/TestSpanAggregation.java PRE-CREATION

>   lang/java/src/test/java/org/apache/avro/ipc/trace/TestSpanTraceFormation.java PRE-CREATION

>   share/schemas/org/apache/avro/ipc/trace/avroTrace.avdl PRE-CREATION 
>   share/schemas/org/apache/avro/ipc/trace/avroTrace.avpr PRE-CREATION 
> 
> Diff: http://review.cloudera.org/r/390/diff
> 
> 
> Testing
> -------
> 
> Several unit-tests are present in this review. See testTraceAndCollection() in TestEndToEndTracing.java
for a clear example of the creation and subsequent aggregation of trace data.
> 
> 
> Thanks,
> 
> Patrick
> 
>


Mime
View raw message