[jira] [Commented] (HTRACE-362) Apache KUDU Span receiver implementation for Apache HTrace

2016-08-23 Thread Nisala Mendis (JIRA) 

[ 
https://issues.apache.org/jira/browse/HTRACE-362?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15434120#comment-15434120
 ] 

Nisala Mendis commented on HTRACE-362:
--

Hi Colin,

I will do the changes to remove opening the table for each span we receive. 
Make sure the table is  properly closed.

{code}
sbuilder.clear()
.setTraceId(span.getSpanId().getHigh())
.setStart(span.getStartTimeMillis())
.setStop(span.getStopTimeMillis())
.setSpanId(span.getSpanId().getLow())
.setProcessId(span.getTracerId())
.setDescription(span.getDescription());

if (span.getParents().length == 0) {
  sbuilder.setParentId(0);
} else if (span.getParents().length > 0) {
  sbuilder.setParentId(span.getParents()[0].getLow());
  if (span.getParents().length > 1) {
LOG.error("error: HBaseSpanReceiver does not support spans " +
"with multiple parents.  Ignoring multiple parents for " +
span);
  }
{code}
I followed the approach which is used in HBase receiver.  If you notice SpanId 
is separated to span id and trace id here. Process is used as tracer id. This 
is the same logic I have followed for persisting parents. I have only persisted 
parent at 0. column_span_parent column I used for convenience when at Span  
viewer, kudu scanner could retrieve root span easily. ( check for root spans ) 
Please check span viewer classes as well. If you can point to kudu binaries you 
can excute unit tests which are currently being disabled.  I have wrote test 
span receiver as well as span viewer.

Regards
Nisala

> Apache KUDU Span receiver implementation for Apache HTrace
> --
>
> Key: HTRACE-362
> URL: https://issues.apache.org/jira/browse/HTRACE-362
> Project: HTrace
>  Issue Type: Bug
>Reporter: Nisala Mendis
>Assignee: Nisala Mendis
> Attachments: HTRACE-362.001.patch, HTRACE-362.002.patch, 
> HTRACE-362.003.patch, HTRACE-362.004.patch, HTRACE-362.005.patch, 
> kudu_receiver.patch, kudu_span_receiver_basic_design.pdf
>
>
> Implementation should be carried two components as span receiver that writes 
> kudu back end and span viewer to view written span/traces.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

[jira] [Commented] (HTRACE-362) Apache KUDU Span receiver implementation for Apache HTrace

2016-08-23 Thread Todd Lipcon (JIRA) 

[ 
https://issues.apache.org/jira/browse/HTRACE-362?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15433794#comment-15433794
 ] 

Todd Lipcon commented on HTRACE-362:


Yes, I recommend keeping a KuduTable and KuduSession open for longer periods of 
time

> Apache KUDU Span receiver implementation for Apache HTrace
> --
>
> Key: HTRACE-362
> URL: https://issues.apache.org/jira/browse/HTRACE-362
> Project: HTrace
>  Issue Type: Bug
>Reporter: Nisala Mendis
>Assignee: Nisala Mendis
> Attachments: HTRACE-362.001.patch, HTRACE-362.002.patch, 
> HTRACE-362.003.patch, HTRACE-362.004.patch, HTRACE-362.005.patch, 
> kudu_receiver.patch, kudu_span_receiver_basic_design.pdf
>
>
> Implementation should be carried two components as span receiver that writes 
> kudu back end and span viewer to view written span/traces.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

[jira] [Commented] (HTRACE-362) Apache KUDU Span receiver implementation for Apache HTrace

2016-08-23 Thread Colin P. McCabe (JIRA) 

[ 
https://issues.apache.org/jira/browse/HTRACE-362?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15433787#comment-15433787
 ] 

Colin P. McCabe commented on HTRACE-362:


Thanks for the confirmation.  While I've got you here :) can we keep the 
KuduTable open while the SpanReceiver is open?  The current patch reopens it on 
each span received, which seems like quite a waste.

> Apache KUDU Span receiver implementation for Apache HTrace
> --
>
> Key: HTRACE-362
> URL: https://issues.apache.org/jira/browse/HTRACE-362
> Project: HTrace
>  Issue Type: Bug
>Reporter: Nisala Mendis
>Assignee: Nisala Mendis
> Attachments: HTRACE-362.001.patch, HTRACE-362.002.patch, 
> HTRACE-362.003.patch, HTRACE-362.004.patch, HTRACE-362.005.patch, 
> kudu_receiver.patch, kudu_span_receiver_basic_design.pdf
>
>
> Implementation should be carried two components as span receiver that writes 
> kudu back end and span viewer to view written span/traces.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

[jira] [Commented] (HTRACE-362) Apache KUDU Span receiver implementation for Apache HTrace

2016-08-23 Thread Todd Lipcon (JIRA) 

[ 
https://issues.apache.org/jira/browse/HTRACE-362?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15433660#comment-15433660
 ] 

Todd Lipcon commented on HTRACE-362:


bq. I'm not sure I follow completely. Are you saying that the cloudera repo in 
the pom.xml is required for now?

Yea we're still working on getting our artifacts up into the ASF maven 
repository, sorry for the delay on that.

> Apache KUDU Span receiver implementation for Apache HTrace
> --
>
> Key: HTRACE-362
> URL: https://issues.apache.org/jira/browse/HTRACE-362
> Project: HTrace
>  Issue Type: Bug
>Reporter: Nisala Mendis
>Assignee: Nisala Mendis
> Attachments: HTRACE-362.001.patch, HTRACE-362.002.patch, 
> HTRACE-362.003.patch, HTRACE-362.004.patch, HTRACE-362.005.patch, 
> kudu_receiver.patch, kudu_span_receiver_basic_design.pdf
>
>
> Implementation should be carried two components as span receiver that writes 
> kudu back end and span viewer to view written span/traces.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)