一.崩潰原因

搭建的是一個hadoop測試集群，所以將數據備份參數設置為dfs.replication=1,這樣如果有一臺datanode損壞的話，數據就會失去。但不幸的是，剛好就有一臺機器由于負載過高，導致數據操壞。進而后面需要重啟整個hadoop集群,重啟后啟動namenode啟動不了。報如下錯誤:

?

Java代碼 ?

FSNamesystem?initialization?failed?saveLeases?found?path????/tmp/xxx/aaa.txt?but?no?matching?entry?in?namespace.??

FSNamesystem initialization failed saveLeases found path /tmp/xxx/aaa.txt but no matching entry in namespace.

?

二.修復namenode?

?

hadoop 集群崩潰了. 導致namenode啟動不了.

?

1. 刪除 namenode主節點的metadata配置目錄

rm -fr /data/hadoop-tmp/hadoop-hadoop/dfs/name

?

2. 啟動secondnamenode

使用start-all.sh命令啟動secondnamenode,namenode的啟動不了不管

?

3. 從secondnamenode恢復

使用命令: hadoop namenode -importCheckpoint

?

?

恢復過程中，發現數據文件有些已經損壞(因為dfs.replication=1),所以一直無法退出安全模式(safemode),一直報如下提示:

?

Java代碼 ?

The?ratio?of?reported?blocks?0.8866?has?not?reached?the?threshold?0.9990.?Safe?mode?will?be?turned?off?automatically.??

The ratio of reported blocks 0.8866 has not reached the threshold 0.9990. Safe mode will be turned off automatically.

?

?

4.強制退出safemode

?

?

Java代碼 ?

hadoop?dfsadmin?-safemode?leave??

hadoop dfsadmin -safemode leave

?

最后啟動成功，查看hdfs網頁報警告信息:

?

?

Java代碼 ?

WARNING?:?There?are?about?257?missing?blocks.?Please?check?the?log?or?run?fsck.??

WARNING : There are about 257 missing blocks. Please check the log or run fsck.

?

?

5.檢查損壞的hdfs文件列表

?

使用命令可以打印出損壞的文件列表:?

?

Java代碼 ?

./hadoop?fsck?/??

./hadoop fsck /

?打印結果:

?

?

?

Java代碼 ?

/user/hive/warehouse/pay_consume_orgi/dt=2011-06-28/consume_2011-06-28.sql:?MISSING?1?blocks?of?total?size?1250990?B.. ??

/user/hive/warehouse/pay_consume_orgi/dt=2011-06-29/consume_2011-06-29.sql:?CORRUPT?block?blk_977550919055291594 ??

??

/user/hive/warehouse/pay_consume_orgi/dt=2011-06-29/consume_2011-06-29.sql:?MISSING?1?blocks?of?total?size?1307147?B..................Status:?CORRUPT ??

?Total?size:????235982871209?B ??

?Total?dirs:????1213??

?Total?files:???1422??

?Total?blocks?(validated):??????4550?(avg.?block?size?51864367?B) ??

??******************************** ??

??CORRUPT?FILES:????????277??

??MISSING?BLOCKS:???????509??

??MISSING?SIZE:?????????21857003415?B ??

??CORRUPT?BLOCKS:???????509??

??********************************??

/user/hive/warehouse/pay_consume_orgi/dt=2011-06-28/consume_2011-06-28.sql: MISSING 1 blocks of total size 1250990 B.. /user/hive/warehouse/pay_consume_orgi/dt=2011-06-29/consume_2011-06-29.sql: CORRUPT block blk_977550919055291594/user/hive/warehouse/pay_consume_orgi/dt=2011-06-29/consume_2011-06-29.sql: MISSING 1 blocks of total size 1307147 B..................Status: CORRUPTTotal size: 235982871209 BTotal dirs: 1213Total files: 1422Total blocks (validated): 4550 (avg. block size 51864367 B)********************************CORRUPT FILES: 277MISSING BLOCKS: 509MISSING SIZE: 21857003415 BCORRUPT BLOCKS: 509********************************

沒有冗余備份,只能刪除損壞的文件,使用命令:

Java代碼 ?

./hadoop?fsck?--delete??

./hadoop fsck --delete

?

?

三.總結

?

一定需要將你的secondnamenode及namenode分開在不同兩臺機器運行，增加namenode的容錯性。以便在集群崩潰時可以從secondnamenode恢復數據.

轉載于:https://www.cnblogs.com/JohnLiang/archive/2011/11/10/2244572.html

總結

以上是生活随笔為你收集整理的hadoop集群崩溃恢复记录的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。