[jira] [Commented] (OAK-6807) Query Recorder

["Thomas Mueller (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/OAK-6807?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16201942#comment-16201942 ] 

Thomas Mueller commented on OAK-6807:
-------------------------------------

First patch (not tested yet). Ideas:

* Uses a new logger, with level "debug" / "trace" (so not enabled by default)
* At level "debug", new queries are logged, plus if they are executed often (after 100 times, 200 times,...)
* Uses at most 5000 queries are kept in memory. If too many, all entries are logged and evicted
* At level "trace", the raw query is logged, and each time it is executed
* At level "debug", queries are simplified, meaning string literals are replaced with 'x' and paths truncated to 2 elements - that way, only the "shape" of the query is kept

[~chetanm] as you have asked for this, maybe you want to quickly review. This is work-in-progress, I will be able to continue in about one week.

{noformat}
### Eclipse Workspace Patch 1.0
#P oak-core
Index: src/main/java/org/apache/jackrabbit/oak/query/stats/QueryStatsData.java
===================================================================
--- src/main/java/org/apache/jackrabbit/oak/query/stats/QueryStatsData.java	(revision 1811333)
+++ src/main/java/org/apache/jackrabbit/oak/query/stats/QueryStatsData.java	(working copy)
@@ -140,6 +140,7 @@
         long time;
         
         public void execute(long nanos) {
+            QueryRecorder.record(query, internal);
             executeCount++;
             lastExecutedMillis = System.currentTimeMillis();
             time += nanos;
Index: src/main/java/org/apache/jackrabbit/oak/query/stats/QueryRecorder.java
===================================================================
--- src/main/java/org/apache/jackrabbit/oak/query/stats/QueryRecorder.java	(nonexistent)
+++ src/main/java/org/apache/jackrabbit/oak/query/stats/QueryRecorder.java	(working copy)
@@ -0,0 +1,126 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.jackrabbit.oak.query.stats;
+
+import java.io.BufferedReader;
+import java.io.FileReader;
+import java.io.IOException;
+import java.io.LineNumberReader;
+import java.util.HashMap;
+import java.util.Map.Entry;
+import java.util.TreeSet;
+
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+public class QueryRecorder {
+
+    private static final Logger LOG = LoggerFactory.getLogger(QueryRecorder.class);
+    private static final HashMap<String, Integer> RECORD_QUERIES_MAP = new HashMap<String, Integer>();
+    private final static int QUERIES_MAX =  Integer.getInteger("oak.query.record", 5000);
+    
+    public static void main(String... args) throws IOException {
+        LineNumberReader reader = new LineNumberReader(
+                new BufferedReader(new FileReader(args[0])));
+        TreeSet<String> sorted = new TreeSet<String>();
+        int lineCount = 0;
+        while(true) {
+            String line = reader.readLine();
+            if(line == null) {
+                break;
+            }
+            sorted.add(simplify(line));
+            lineCount++;
+        }
+        reader.close();
+        for(String s : sorted) {
+            System.out.println(s);
+        }
+        System.out.println("sorted: " + sorted.size() + " original: " + lineCount);
+    }
+    
+    public static void record(String query, boolean internal) {
+        if (internal) {
+            return;
+        }
+        if (!LOG.isDebugEnabled()) {
+            return;
+        }
+        if (LOG.isTraceEnabled()) {
+            LOG.trace("query {}", query);
+            return;
+        }
+        query = simplify(query);
+        record(query);
+    }
+    
+    private static String simplify(String query) {
+        query = query.replaceAll("'[^']*'", "'x'");
+        query = query.replaceAll("ISDESCENDANTNODE(\\[[^]]\\])", "ISDESCENDANTNODE('x')");
+        int pathIndex = query.indexOf("/jcr:root/");
+        if (pathIndex >= 0) {
+            int end = getFirstOccurance(query, pathIndex,
+                    " ", "/element(", "/text(", "/*", "/(", "/jcr:deref(");
+            String path = query.substring(pathIndex + "/jcr:root/".length());
+            int first = path.indexOf('/');
+            if (first > 0) {
+                first = path.indexOf('/', first + 1);
+                if (first > 0) {
+                    path = path.substring(0, first);
+                }
+            }
+            String newQuery = query.substring(0, pathIndex) + "/jcr:root/" + path + 
+                    "..." + query.substring(end, query.length());
+            System.out.println(query);
+            System.out.println(newQuery);
+            System.out.println();
+            query = newQuery;
+        }
+        return query;
+    }
+    
+    static int getFirstOccurance(String text, int start, String... strings) {
+        int first = text.length();
+        for(String s : strings) {
+            int index = text.indexOf(s, start);
+            if (index > 0 && index < first) {
+                first = index;
+            }
+        }
+        return first;
+    }
+    
+    private static synchronized void record(String query) {
+        if (RECORD_QUERIES_MAP.size() > QUERIES_MAX) {
+            for(Entry<String, Integer> e : RECORD_QUERIES_MAP.entrySet()) {
+                log(e.getKey(), e.getValue());
+            }
+            RECORD_QUERIES_MAP.clear();
+        }
+        Integer count = RECORD_QUERIES_MAP.get(query);
+        count = count == null ? 1 : count + 1;
+        RECORD_QUERIES_MAP.put(query, count);
+        if (count == 0 || count % 100 == 0) {
+            log(query, count);
+        }
+    }
+
+    private static void log(String query, int count) {
+        LOG.debug("query {} count {}", query, count);
+    }
+
+}
{noformat}

> Query Recorder
> --------------
>
>                 Key: OAK-6807
>                 URL: https://issues.apache.org/jira/browse/OAK-6807
>             Project: Jackrabbit Oak
>          Issue Type: Improvement
>          Components: query
>            Reporter: Thomas Mueller
>            Assignee: Thomas Mueller
>             Fix For: 1.8
>
>
> In order to manage indexes (e.g. find out which indexes are no longer needed, which properties don't need to be indexed any longer), we have an easy way to log all executed queries / query plans. 
> Each entry only needs to be logged once (logging multiple times is OK, but ensure it's not logged to often). Different log levels can be used (e.g. log level "TRACE" logs more data, "DEBUG" less). For "DEBUG" level, overhead of logging should be minimal, so this can be kept enabled for a long time.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)