Cannot install rhdfs at my EC2 unbuntu instances

[gz5427]

Hi,

I am trying to install "rhdfs" at my EC2 unbuntu instances. I first installed the rhdfs dependent packages: such as rJava, Plyr, rmr2. But when I tryied to run the following command to install rhdfs,

sudo R CMD INSTALL /data/tarfiles/rhdfs_1.0.8.tar.gz

I got the following error message:

** testing if installed package can be loaded
Error : .onLoad failed in loadNamespace() for 'rhdfs', details:
call: fun(libname, pkgname)
error: Environment variable HADOOP_CMD must be set before loading package rhdfs
Error: loading failed

I tried to open R and run "Sys.getenv("HADOOP_CMD")". It does give me the right HADOOP HOME. So the error message is really weird for me.

Can anybody help?

Max

[piccolbo]

sudo resets most of the environment. You can use sudo -E but it's
considered an insecure option so please read man sudo before acting.

Antonio

On Wed, Jan 21, 2015 at 8:38 AM, gz5427 notifications@github.com wrote:

Hi,
I am trying to install "rhdfs" at my EC2 unbuntu instances. I first
installed the rhdfs dependent packages: such as rJava, Plyr, rmr2. But when
I tryied to run the following command to install rhdfs,

sudo R CMD INSTALL /data/tarfiles/rhdfs_1.0.8.tar.gz
I got the following error message:

** testing if installed package can be loaded
Error : .onLoad failed in loadNamespace() for 'rhdfs', details:
call: fun(libname, pkgname)
error: Environment variable HADOOP_CMD must be set before loading package
rhdfs
Error: loading failed

I tried to open R and run "Sys.getenv("HADOOP_CMD")". It does give me the
right HADOOP HOME. So the error message is really weird for me.

Can anybody help?

Max

—
Reply to this email directly or view it on GitHub
#14.

[gz5427]

Hi Antonio,

Could you tell me what other environmental parameters I need to reset?
Right now, I put the following in my .bashrc:

export HADOOP_CMD=/data/hadoop-2.6.0
export
HADOOP_STREAMING_JAR=/data/hadoop-2.6.0/share/hadoop/tools/lib/hadoop-streaming-2.6.0.jar
export
HADOOP_STREAMING=/data/hadoop-2.6.0/share/hadoop/tools/lib/hadoop-streaming-2.6.0.jar

And when I opened R, these variables were read in R as well.

Thanks,
Max

On Wed, Jan 21, 2015 at 11:58 AM, Antonio Piccolboni <
notifications@github.com> wrote:

sudo resets most of the environment. You can use sudo -E but it's
considered an insecure option so please read man sudo before acting.

Antonio

On Wed, Jan 21, 2015 at 8:38 AM, gz5427 notifications@github.com wrote:

Hi,
I am trying to install "rhdfs" at my EC2 unbuntu instances. I first
installed the rhdfs dependent packages: such as rJava, Plyr, rmr2. But
when
I tryied to run the following command to install rhdfs,

sudo R CMD INSTALL /data/tarfiles/rhdfs_1.0.8.tar.gz
I got the following error message:

** testing if installed package can be loaded
Error : .onLoad failed in loadNamespace() for 'rhdfs', details:
call: fun(libname, pkgname)
error: Environment variable HADOOP_CMD must be set before loading
package
rhdfs
Error: loading failed

I tried to open R and run "Sys.getenv("HADOOP_CMD")". It does give me
the
right HADOOP HOME. So the error message is really weird for me.

Can anybody help?

Max

—
Reply to this email directly or view it on GitHub
#14.

—
Reply to this email directly or view it on GitHub
#14 (comment)
.

[piccolbo]

You need to set or reset no additional variables. You need to learn how
sudo works or avoid it.

Antonio:~ antonio$ env |grep HADOOP_STRE

HADOOP_STREAMING=/Users/antonio/Downloads/hadoop-2.2.0//share/hadoop/tools/lib/hadoop-streaming-2.2.0.jar

Antonio:~ antonio$ sudo env |grep HADOOP_STRE

Antonio:~ antonio$ sudo -E env |grep HADOOP_STRE

HADOOP_STREAMING=/Users/antonio/Downloads/hadoop-2.2.0//share/hadoop/tools/lib/hadoop-streaming-2.2.0.jar

Or try sudo R and do a sys.getenv at the R prompt . Or read the manual
before using it.
Particularly with dangerous command like sudo, read the manual, don't
guess. Don't use sudo -E because I said so, read the manual first. It has
security implications.

On Wed, Jan 21, 2015 at 6:41 PM, gz5427 notifications@github.com wrote:

Hi Antonio,

Could you tell me what other environmental parameters I need to reset?
Right now, I put the following in my .bashrc:

export HADOOP_CMD=/data/hadoop-2.6.0
export

HADOOP_STREAMING_JAR=/data/hadoop-2.6.0/share/hadoop/tools/lib/hadoop-streaming-2.6.0.jar
export

HADOOP_STREAMING=/data/hadoop-2.6.0/share/hadoop/tools/lib/hadoop-streaming-2.6.0.jar

And when I opened R, these variables were read in R as well.

Thanks,
Max

On Wed, Jan 21, 2015 at 11:58 AM, Antonio Piccolboni <
notifications@github.com> wrote:

sudo resets most of the environment. You can use sudo -E but it's
considered an insecure option so please read man sudo before acting.

Antonio

On Wed, Jan 21, 2015 at 8:38 AM, gz5427 notifications@github.com
wrote:

Hi,
I am trying to install "rhdfs" at my EC2 unbuntu instances. I first
installed the rhdfs dependent packages: such as rJava, Plyr, rmr2. But
when
I tryied to run the following command to install rhdfs,

sudo R CMD INSTALL /data/tarfiles/rhdfs_1.0.8.tar.gz
I got the following error message:

** testing if installed package can be loaded
Error : .onLoad failed in loadNamespace() for 'rhdfs', details:
call: fun(libname, pkgname)
error: Environment variable HADOOP_CMD must be set before loading
package
rhdfs
Error: loading failed

I tried to open R and run "Sys.getenv("HADOOP_CMD")". It does give me
the
right HADOOP HOME. So the error message is really weird for me.

Can anybody help?

Max

—
Reply to this email directly or view it on GitHub
#14.

—
Reply to this email directly or view it on GitHub
<
#14 (comment)

.

—
Reply to this email directly or view it on GitHub.

[gz5427]

Antonio,

I simply log-in as super user and run the R CMD INSTALL. And it works.

I am just wondering why this package differ with us so much. Particularly
the error message is not very precise! For example if I am install it as
the "ubuntu", and the "HADOOP_CMD" and the "HADOOP_STREAMING" has been set
in the current R session. Why it won't work? I think it is very
confusing for lots of R users!

Well, I really appreciate your support on this, explaining the "env"
Max

On Wed, Jan 21, 2015 at 11:19 PM, Antonio Piccolboni <
notifications@github.com> wrote:

You need to set or reset no additional variables. You need to learn how
sudo works or avoid it.

Antonio:~ antonio$ env |grep HADOOP_STRE

HADOOP_STREAMING=/Users/antonio/Downloads/hadoop-2.2.0//share/hadoop/tools/lib/hadoop-streaming-2.2.0.jar


Antonio:~ antonio$ sudo env |grep HADOOP_STRE

Antonio:~ antonio$ sudo -E env |grep HADOOP_STRE

HADOOP_STREAMING=/Users/antonio/Downloads/hadoop-2.2.0//share/hadoop/tools/lib/hadoop-streaming-2.2.0.jar

Or try sudo R and do a sys.getenv at the R prompt . Or read the manual
before using it.
Particularly with dangerous command like sudo, read the manual, don't
guess. Don't use sudo -E because I said so, read the manual first. It has
security implications.

On Wed, Jan 21, 2015 at 6:41 PM, gz5427 notifications@github.com wrote:

Hi Antonio,

Could you tell me what other environmental parameters I need to reset?
Right now, I put the following in my .bashrc:

export HADOOP_CMD=/data/hadoop-2.6.0
export

HADOOP_STREAMING_JAR=/data/hadoop-2.6.0/share/hadoop/tools/lib/hadoop-streaming-2.6.0.jar

export

HADOOP_STREAMING=/data/hadoop-2.6.0/share/hadoop/tools/lib/hadoop-streaming-2.6.0.jar

And when I opened R, these variables were read in R as well.

Thanks,
Max

On Wed, Jan 21, 2015 at 11:58 AM, Antonio Piccolboni <
notifications@github.com> wrote:

sudo resets most of the environment. You can use sudo -E but it's
considered an insecure option so please read man sudo before acting.

Antonio

On Wed, Jan 21, 2015 at 8:38 AM, gz5427 notifications@github.com
wrote:

Hi,
I am trying to install "rhdfs" at my EC2 unbuntu instances. I first
installed the rhdfs dependent packages: such as rJava, Plyr, rmr2.
But
when
I tryied to run the following command to install rhdfs,

sudo R CMD INSTALL /data/tarfiles/rhdfs_1.0.8.tar.gz
I got the following error message:

** testing if installed package can be loaded
Error : .onLoad failed in loadNamespace() for 'rhdfs', details:
call: fun(libname, pkgname)
error: Environment variable HADOOP_CMD must be set before loading
package
rhdfs
Error: loading failed

I tried to open R and run "Sys.getenv("HADOOP_CMD")". It does give
me
the
right HADOOP HOME. So the error message is really weird for me.

Can anybody help?

Max

—
Reply to this email directly or view it on GitHub
#14.

—
Reply to this email directly or view it on GitHub
<

#14 (comment)

.

—
Reply to this email directly or view it on GitHub.

—
Reply to this email directly or view it on GitHub
#14 (comment)
.

[msolano00]

Hi @gz5427,

You can set the environment variable for the R session itself by:

Sys.setenv(HADOOP_CMD="/usr/bin/hadoop")