[
https://issues.apache.org/jira/browse/HBASE-21926?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16794138#comment-16794138
]
Sean Busbey commented on HBASE-21926:
-------------------------------------
I won't have time to make my desired changes until early next week earliest.
$dayjob priorities changed for this last week.
bq. I'm going to ignore further ImportOrder checkstyle warnings, because it
doesn't seem to matter where the import is placed, there is still a warning.
Tired of guessing what is the right answer. As this is the only checkstyle
warning the patch looks good IMHO.
FWIW, there's supposed to be both an IDE formatter config and a write up
somewhere explaining the current configuration. [we define 3 import
"groups"|https://github.com/apache/hbase/blob/master/hbase-checkstyle/src/main/resources/hbase/checkstyle.xml#L71].
which I think means imports should be "things that aren't thirdparty/shaded
first in alpha order", followed by "thirdparty in alpha order", followed by
"shaded in alpha order".
If I understand the config correctly, in ProfileServlet this line should be
after all of the other imports:
{code}
34 import org.apache.hbase.thirdparty.com.google.common.base.Joiner;
{code}
and in ProcessUtils this line should be before the slf4j imports
{code}
27 import org.apache.yetus.audience.InterfaceAudience;
{code}
> Profiler servlet
> ----------------
>
> Key: HBASE-21926
> URL: https://issues.apache.org/jira/browse/HBASE-21926
> Project: HBase
> Issue Type: New Feature
> Components: master, Operability, regionserver
> Reporter: Andrew Purtell
> Assignee: Andrew Purtell
> Priority: Major
> Fix For: 3.0.0, 1.5.0, 2.3.0
>
> Attachments: 1.png, 2.png, 3.png, 4.png, HBASE-21926-branch-1.patch,
> HBASE-21926-branch-1.patch, HBASE-21926-branch-1.patch, HBASE-21926.patch,
> HBASE-21926.patch, HBASE-21926.patch
>
>
> HIVE-20202 describes how Hive added a web endpoint for online in production
> profiling based on async-profiler. The endpoint was added as a servlet to
> httpserver and supports retrieval of flamegraphs compiled from the profiler
> trace. Async profiler
> ([https://github.com/jvm-profiling-tools/async-profiler] ) can also profile
> heap allocations, lock contention, and HW performance counters in addition to
> CPU.
> The profiling overhead is pretty low and is safe to run in production. The
> async-profiler project measured and describes CPU and memory overheads on
> these issues:
> [https://github.com/jvm-profiling-tools/async-profiler/issues/14] and
> [https://github.com/jvm-profiling-tools/async-profiler/issues/131]
> We have an httpserver based servlet stack so we can use HIVE-20202 as an
> implementation template for a similar feature for HBase daemons. Ideally we
> achieve these requirements:
> * Retrieve flamegraph SVG generated from latest profile trace.
> * Online enable and disable of profiling activity. (async-profiler does not
> do instrumentation based profiling so this should not cause the code gen
> related perf problems of that other approach and can be safely toggled on and
> off while under production load.)
> * CPU profiling.
> * ALLOCATION profiling.
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
