- added xml output in IndexControlURLs to get the storage page of index

dump commands
- adjusted the apicall.sh script to get the downloaded text as output to
stdout which is necessary to parse the content out of it
- added indexdump.sh script which creates a solr dump and prints out the
storage path for the index dump
- added synchronization to the Fulltext class to prevent that data is
stored to a non-existing solr index while this index is disabled during
the storage of the dump

bin/apicall.sh

@@ -4,9 +4,9 @@ port=$(grep ^port= ../DATA/SETTINGS/yacy.conf |cut -d= -f2)
 pw=$(grep ^adminAccountBase64MD5= ../DATA/SETTINGS/yacy.conf |cut -d= -f2)
 
 if which curl &>/dev/null; then
-  curl -s --header "Authorization: realm=$pw" "http://127.0.0.1:$port/$1" > /dev/null
+  curl -s --header "Authorization: realm=$pw" "http://127.0.0.1:$port/$1"
 elif which wget &>/dev/null; then
-  wget -q -t 1 --timeout=5 --header "Authorization: realm=$pw" "http://127.0.0.1:$port/$1" -O /dev/null
+  wget -q -t 1 --timeout=120 --header "Authorization: realm=$pw" "http://127.0.0.1:$port/$1" -O -
 else
   exit 1
 fi

bin/indexdump.sh

@@ -0,0 +1,3 @@
+#!/bin/bash
+cd "`dirname $0`"
+./apicall0.sh "/IndexControlURLs_p.xml?indexdump=" | awk '/<dumpfile>/{ gsub("<dumpfile>","" );gsub("<\/dumpfile>","" ); print $0 }' | awk '{print $1}';

htroot/IndexControlURLs_p.xml

@@ -0,0 +1,25 @@
+<?xml version="1.0"?>
+<data>
+    #(statisticslines)#::
+    <domains>
+      #{domains}#
+      <domain>
+        <host>#[domain]#</host>
+        <count>#[count]#</count>
+      </domain>
+      #{/domains}#
+    </domains>
+    #(/statisticslines)#
+    #(indexdump)#::
+    <dumpfile>#[dumpfile]#</dumpfile>::
+    #(/indexdump)#
+    #(urlhashsimilar)#::
+    <urls>
+    #{rows}# 
+    #{cols}#
+    <urlhash>#[urlHash]#</urlhash>
+    #{/cols}#
+    #{/rows}#
+    </urls>
+    #(/urlhashsimilar)#
+</data>

source/net/yacy/search/index/Fulltext.java

@@ -293,10 +293,12 @@ public final class Fulltext implements Iterable<byte[]> {
         if (this.connectedSolr()) {
 	        try {
 	        	if (this.urlIndexFile != null) this.urlIndexFile.remove(idb);
-	            SolrDocument sd = this.solr.get(id);
-	            if (sd == null || this.solrScheme.getDate(sd, YaCySchema.last_modified).before(this.solrScheme.getDate(doc, YaCySchema.last_modified))) {
-		            this.solr.add(doc);
-	            }
+	        	synchronized (this.solr) {
+	        	    SolrDocument sd = this.solr.get(id);
+	        	    if (sd == null || this.solrScheme.getDate(sd, YaCySchema.last_modified).before(this.solrScheme.getDate(doc, YaCySchema.last_modified))) {
+	        	        this.solr.add(doc);
+	        	    }
+	        	}
 	        } catch (SolrException e) {
 	            throw new IOException(e.getMessage(), e);
 	        }
@@ -334,10 +336,12 @@ public final class Fulltext implements Iterable<byte[]> {
         if (this.connectedSolr()) {
 	        try {
 	        	if (this.urlIndexFile != null) this.urlIndexFile.remove(idb);
-	            SolrDocument sd = this.solr.get(id);
-	            if (sd == null || (new URIMetadataNode(sd)).isOlder(row)) {
-	            	this.solr.add(getSolrScheme().metadata2solr(row));
-	            }
+                synchronized (this.solr) {
+                    SolrDocument sd = this.solr.get(id);
+                    if (sd == null || (new URIMetadataNode(sd)).isOlder(row)) {
+                        this.solr.add(getSolrScheme().metadata2solr(row));
+                    }
+                }
 	        } catch (SolrException e) {
 	            throw new IOException(e.getMessage(), e);
 	        }
@@ -365,7 +369,9 @@ public final class Fulltext implements Iterable<byte[]> {
     public boolean remove(final byte[] urlHash) {
         if (urlHash == null) return false;
         try {
-            this.solr.delete(ASCII.String(urlHash));
+            synchronized (this.solr) {
+                this.solr.delete(ASCII.String(urlHash));
+            }
         } catch (final Throwable e) {
             Log.logException(e);
         }
@@ -470,17 +476,19 @@ public final class Fulltext implements Iterable<byte[]> {
         EmbeddedSolrConnector esc = (EmbeddedSolrConnector) this.solr.getSolr0();
         int commitWithin = esc.getCommitWithinMs();
         File storagePath = esc.getStoragePath();
-        this.disconnectLocalSolr();
         File zipOut = new File(storagePath.toString() + "_" + GenericFormatter.SHORT_DAY_FORMATTER.format() + ".zip");
-        try {
-            ZIPWriter.zip(storagePath, zipOut);
-        } catch (IOException e) {
-            Log.logException(e);
-        } finally {
+        synchronized (this.solr) {
+            this.disconnectLocalSolr();
             try {
-                this.connectLocalSolr(commitWithin);
+                ZIPWriter.zip(storagePath, zipOut);
             } catch (IOException e) {
                 Log.logException(e);
+            } finally {
+                try {
+                    this.connectLocalSolr(commitWithin);
+                } catch (IOException e) {
+                    Log.logException(e);
+                }
             }
         }
         return zipOut;
@@ -494,16 +502,18 @@ public final class Fulltext implements Iterable<byte[]> {
         EmbeddedSolrConnector esc = (EmbeddedSolrConnector) this.solr.getSolr0();
         int commitWithin = esc.getCommitWithinMs();
         File storagePath = esc.getStoragePath();
-        this.disconnectLocalSolr();
-        try {
-            ZIPReader.unzip(solrDumpZipFile, storagePath);
-        } catch (IOException e) {
-            Log.logException(e);
-        } finally {
+        synchronized (this.solr) {
+            this.disconnectLocalSolr();
             try {
-                this.connectLocalSolr(commitWithin);
+                ZIPReader.unzip(solrDumpZipFile, storagePath);
             } catch (IOException e) {
                 Log.logException(e);
+            } finally {
+                try {
+                    this.connectLocalSolr(commitWithin);
+                } catch (IOException e) {
+                    Log.logException(e);
+                }
             }
         }
     }
@@ -783,7 +793,9 @@ public final class Fulltext implements Iterable<byte[]> {
         // first collect all url hashes that belong to the domain
         assert hosthash.length() == 6;
         // delete in solr
-        this.solr.deleteByQuery(YaCySchema.host_id_s.name() + ":\"" + hosthash + "\"");
+        synchronized (this.solr) {
+            this.solr.deleteByQuery(YaCySchema.host_id_s.name() + ":\"" + hosthash + "\"");
+        }
 
         // delete in old metadata structure
         final ArrayList<String> l = new ArrayList<String>();