ERROR org.apache.hadoop.hdfs.server.namenode.NameNode: java.lang.NumberFormatExc

yuandan

　　namenode莫名奇妙的启动不了，看log：
　　2011-04-19 12:06:59,967 INFO org.apache.hadoop.hdfs.server.common.Storage: Number of files = 11471
2011-04-19 12:07:00,592 INFO org.apache.hadoop.hdfs.server.common.Storage: Number of files under construction = 0
2011-04-19 12:07:00,592 INFO org.apache.hadoop.hdfs.server.common.Storage: Image file of size 1722772 loaded in 0 seconds.
2011-04-19 12:07:00,680 ERROR org.apache.hadoop.hdfs.server.namenode.NameNode: java.lang.NumberFormatException: For input string: "13031^@^@^@^@^@^@^@^@"
        at java.lang.NumberFormatException.forInputString(NumberFormatException.java:48)
        at java.lang.Long.parseLong(Long.java:419)
        at java.lang.Long.parseLong(Long.java:468)
        at org.apache.hadoop.hdfs.server.namenode.FSEditLog.readLong(FSEditLog.java:1470)
        at org.apache.hadoop.hdfs.server.namenode.FSEditLog.loadFSEdits(FSEditLog.java:797)
        at org.apache.hadoop.hdfs.server.namenode.FSImage.loadFSEdits(FSImage.java:1034)
        at org.apache.hadoop.hdfs.server.namenode.FSImage.loadFSImage(FSImage.java:845)
        at org.apache.hadoop.hdfs.server.namenode.FSImage.recoverTransitionRead(FSImage.java:379)
        at org.apache.hadoop.hdfs.server.namenode.FSDirectory.loadFSImage(FSDirectory.java:99)
        at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.initialize(FSNamesystem.java:347)
        at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.<init>(FSNamesystem.java:321)
        at org.apache.hadoop.hdfs.server.namenode.NameNode.initialize(NameNode.java:267)
        at org.apache.hadoop.hdfs.server.namenode.NameNode.<init>(NameNode.java:461)
        at org.apache.hadoop.hdfs.server.namenode.NameNode.createNameNode(NameNode.java:1202)
        at org.apache.hadoop.hdfs.server.namenode.NameNode.main(NameNode.java:1211)
　　这是神马意思？？？  我就于是乎翻hadoop mail list的邮件，还真叫我找到了！！！ mail list 果然很强大的说。。下面是连接http://mail-archives.apache.org/mod_mbox/hadoop-common-user/201003.mbox/%3c2986c2f31003041137j3410bed6wab112faf8f7b605c@mail.gmail.com%3e
　　http://mail-archives.apache.org/mod_mbox/hadoop-hdfs-user/201010.mbox/%3cBCCFEB17-8464-466B-BD54-125764974AD5@mlogiciels.com%3e
　　最后还是选择用secondNamenode里的editlog替换掉namenode里的，start-all.sh 后，能正常使用。fsck / 一下，还好没有丢失数据。。。 至今不明白谁家那小谁做了神马操作导致这个情况。。。