hadoop-hdfs-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alex Kozlov <ale...@cloudera.com>
Subject Re: How do I fix a corrupt hdfs file system
Date Tue, 24 Jan 2012 21:11:16 GMT
Hi Steve, It says "Number of data-nodes: 2".  Are all of your datanodes up?

On Tue, Jan 24, 2012 at 11:55 AM, Steve Lewis <lordjoe2000@gmail.com> wrote:

> I did that that is even it said it was corrupt
> On Jan 24, 2012 10:17 AM, "alo alt" <wget.null@googlemail.com> wrote:
>
>> hadoop dfs -fsck /
>>
>> - Alex
>>
>> --
>> Alexander Lorenz
>> http://mapredit.blogspot.com
>>
>> On Jan 24, 2012, at 7:15 PM, Steve Lewis wrote:
>>
>> > When I do an fsck / this is what I get and my jobs are failing with
>>  Could not obtain block
>> >
>> > /user/www/testoutput/part-00000:  Under replicated
>> blk_4592416107180235380_1833133. Target Replicas is 3 but found 2
>> replica(s).
>> > Status: CORRUPT
>> >  Total size:    553069957701 B
>> >  Total dirs:    13319
>> >  Total files:   127777
>> >  Total blocks (validated):      21117 (avg. block size 26190744 B)
>> >   ********************************
>> >   CORRUPT FILES:        3527
>> >   MISSING BLOCKS:       3624
>> >   MISSING SIZE:         72682479091 B
>> >   CORRUPT BLOCKS:       3624
>> >   ********************************
>> >  Minimally replicated blocks:   17493 (82.83847 %)
>> >  Over-replicated blocks:        0 (0.0 %)
>> >  Under-replicated blocks:       17493 (82.83847 %)
>> >  Mis-replicated blocks:         0 (0.0 %)
>> >  Default replication factor:    3
>> >  Average block replication:     1.5938344
>> >  Corrupt blocks:                3624
>> >  Missing replicas:              18864 (56.047775 %)
>> >  Number of data-nodes:          2
>> >  Number of racks:               1
>> >
>> >
>> > The filesystem under path '/' is CORRUPT
>> >
>> >
>> > --
>> > Steven M. Lewis PhD
>> > 4221 105th Ave NE
>> > Kirkland, WA 98033
>> > 206-384-1340 (cell)
>> > Skype lordjoe_com
>> >
>> >
>>
>>

Mime
View raw message