Author: tejasp
Date: Sun Apr 28 01:13:21 2013
New Revision: 1476702
URL: http://svn.apache.org/r1476702
Log:
NUTCH-829 duplicate hadoop temp files
Modified:
nutch/trunk/CHANGES.txt
nutch/trunk/src/java/org/apache/nutch/crawl/Generator.java
Modified: nutch/trunk/CHANGES.txt
URL:
http://svn.apache.org/viewvc/nutch/trunk/CHANGES.txt?rev=1476702&r1=1476701&r2=1476702&view=diff
==============================================================================
--- nutch/trunk/CHANGES.txt (original)
+++ nutch/trunk/CHANGES.txt Sun Apr 28 01:13:21 2013
@@ -2,6 +2,8 @@ Nutch Change Log
(trunk): Current Development
+* NUTCH-829 duplicate hadoop temp files (Mike Baranczak, lewismc, tejasp)
+
* NUTCH-1501 Harmonize behavior of parsechecker and indexchecker (snagel +
lewismc)
* NUTCH-1031 Delegate parsing of robots.txt to crawler-commons (tejasp)
Modified: nutch/trunk/src/java/org/apache/nutch/crawl/Generator.java
URL:
http://svn.apache.org/viewvc/nutch/trunk/src/java/org/apache/nutch/crawl/Generator.java?rev=1476702&r1=1476701&r2=1476702&view=diff
==============================================================================
--- nutch/trunk/src/java/org/apache/nutch/crawl/Generator.java (original)
+++ nutch/trunk/src/java/org/apache/nutch/crawl/Generator.java Sun Apr 28
01:13:21 2013
@@ -492,7 +492,7 @@ public class Generator extends Configure
throws IOException {
Path tempDir = new Path(getConf().get("mapred.temp.dir", ".") +
"/generate-temp-"
- + System.currentTimeMillis());
+ + java.util.UUID.randomUUID().toString());
Path lock = new Path(dbDir, CrawlDb.LOCK_NAME);
FileSystem fs = FileSystem.get(getConf());
@@ -582,7 +582,7 @@ public class Generator extends Configure
if (getConf().getBoolean(GENERATE_UPDATE_CRAWLDB, false)) {
// update the db from tempDir
Path tempDir2 = new Path(getConf().get("mapred.temp.dir", ".") +
"/generate-temp-"
- + System.currentTimeMillis());
+ + java.util.UUID.randomUUID().toString());
job = new NutchJob(getConf());
job.setJobName("generate: updatedb " + dbDir);