hadoop-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Harsh J <ha...@cloudera.com>
Subject Re: fsck only working on namenode
Date Wed, 07 Nov 2012 16:58:11 GMT
While your problem is interesting, you need not use FSCK to get block
IDs of a file, as thats not the right way to fetch it (its a rather
long, should-be-disallowed route). You can leverage the FileSystem API
itself to do that. See FileSystem#getFileBlockLocations(…), i.e.
http://hadoop.apache.org/docs/current/api/org/apache/hadoop/fs/FileSystem.html#getFileBlockLocations(org.apache.hadoop.fs.FileStatus,%20long,%20long)
if you use FileSystem APIs, or see FileContext#listLocatedStatus(…)
i.e. http://hadoop.apache.org/docs/current/api/org/apache/hadoop/fs/FileContext.html#listLocatedStatus(org.apache.hadoop.fs.Path)
if you use FileContext APIs.

Onto your problem though, can you successfully do a `telnet NNHOST
50070` from one of your slave nodes?

On Wed, Nov 7, 2012 at 10:18 PM, Sebastian.Lehrack
<Sebastian.Lehrack@physik.uni-muenchen.de> wrote:
> Hi,
>
> I've installed hadoop 1.0.3 on a cluster of about 25 nodes and till now,
> it's working fine.
> Recently, i had to use fsck in a map-process, which leads to a
> connection refused error.
> I read about this error, that i should check about firewalls and proper
> configfiles etc.
> The command is only working on the namenode.
> If i use the browser for the command, it's working (although also
> refused, but because of webusers permission)
> I can use telnet to connect to the namenode.
> In hdfs-site.conf, i set dfs.http.adress to hostname:50070. I tried
> IP-adress and Hostname. I marked it as final.
> I'm still getting this connecting refused error, when using fsck on a
> node other then the namenode.
>
> Any further suggesting would be great. The fsck command is used to check
> the numbers of block, in which a file is stored on the hdfs. Maybe
> there's another possibility?
>
> Greetings



-- 
Harsh J

Mime
View raw message