Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Hive >> mail # user >> RE: Regarding Hive

Copy link to this message
Re: Regarding Hive

I have Hive running with cygwin and windows, you need to apply this
patch though:


It will work fine except if you use AUX_PARAMs which I still have to
find the time to fix.

The other options for running on windows are:

* Run a Linux virtual machine (e.g.

* Run coLinux (www.colinux.org), which is really like a virtual machine
with linux, but it doesn't require a virtual machine software like
vmware or xen. You will still need to scp the files between the guest
linux machine and the host windows machine, so not as smooth as cygwin.

-- amr

Prasad Chakka wrote:
> Hive does not support cygwin but there are developers that are running
> on Windows using some other system. Hopefully they will chime in
> (cc'ing the apache hive user list).
> Do a 'ps --Afw | grep --I hive' to see all the hive processes on your
> unix system.
> Prasad
> ------------------------------------------------------------------------
> *From: *Haritha Javvadi <[EMAIL PROTECTED]>
> *Date: *Tue, 7 Jul 2009 17:34:28 -0700
> *To: *Prasad Chakka <[EMAIL PROTECTED]>
> *Subject: *Re: Regarding Hive
> Hey,
>              OK.got it.Can you suggest me whats the method involved in
> killing the process. What should be done to kill the process. I think,
> the mistake I have done was: Opened the server through putty,later
> hive and was running a query from two tables using join. It was really
> slow. So I opened another window of the server and again accessed hive
> from that window for another query. This was done on the same server.
> Is that the problem. I have one more question. My local machine has
> Windows XP OS. I read in some documents that hive does not support
> cygwin. Is this information correct.
> Haritha
> On Tue, Jul 7, 2009 at 5:21 PM, Prasad Chakka <[EMAIL PROTECTED]>
> wrote:
>     This is from log:
>     Caused by: java.sql.SQLException: Another instance of Derby may
>     have already booted the database /var/hive/metastore_db.
>     So there must be some other derby/hive process on this machine
>     that has opened this file. You need kill this process or recreate
>     metadata by deleting /var/hive/metastore_db (you will loose all
>     the metadata that you have created). You can check the wiki link I
>     have sent you on how to create a multi-user metastore (either a
>     derby server or a mysql server).
>     Prasad
>     ------------------------------------------------------------------------
>     *From: *Haritha Javvadi <[EMAIL PROTECTED]
>     <http://[EMAIL PROTECTED]> >
>     *Date: *Tue, 7 Jul 2009 17:18:08 -0700
>     *To: *Prasad Chakka <[EMAIL PROTECTED]
>     <http://[EMAIL PROTECTED]> >
>     *Subject: *Re: Regarding Hive
>     Hey Prasad,
>                            I am sorry for that.I have attached the
>     hive.log.Please check it out.
>     Haritha
>     On Tue, Jul 7, 2009 at 5:13 PM, Prasad Chakka
>     <[EMAIL PROTECTED] <http://[EMAIL PROTECTED]> > wrote:
>         Logs are not in hdfs but on local file system where you are
>         executing hive client.
>         ------------------------------------------------------------------------
>         *From: *Haritha Javvadi <[EMAIL PROTECTED]
>         <http://[EMAIL PROTECTED]>
>          <http://[EMAIL PROTECTED]> >
>         *Date: *Tue, 7 Jul 2009 17:10:16 -0700
>         *To: *Prasad Chakka <[EMAIL PROTECTED]
>         <http://[EMAIL PROTECTED]>  <http://[EMAIL PROTECTED]> >
>         *Subject: *Re: Regarding Hive
>         Hey Prasad,
>                           I tried to get the hive.log but it didnt
>         show up any hive.log files in hdfs. It showed up the following:
>         [root@dev1 hadoop-0.18.3]# bin/hadoop dfs -ls /tmp
>         Found 10 items
>         drwxr-xr-x   - root supergroup          0 2009-07-07 07:10