[CentOS] ssh remote execute awk problem

[Nico Kadel-Garcia]

On Thu, Mar 17, 2011 at 9:31 AM, Nico Kadel-Garcia <nkadel at gmail.com> wrote:
> On Thu, Mar 17, 2011 at 6:58 AM, John Doe <jdmls at yahoo.com> wrote:
>> From: sync <jiannma at gmail.com>
>>>There is a problem on executing awk command to remote servers;
>>>    1. #!/bin/sh
>>>    2.
>>>    3. _CMD="ssh root at localhost"
>>>    4.
>>>    5. cpu_num="$($_CMD awk '/processor/{count["proc"]++}; END{print
>>>count["proc"]}' /proc/cpuinfo)“
>>>    6.
>>>    7. echo $cpu_num
>>
>> Quote the whole command and backslash the command quotes...
>> ssh root at localhost "awk '/processor/{count[\"proc\"]++}; END{print
>> count[\"proc\"] } ' /proc/cpuinfo"
>>
>> JD
>
> Or do the processing locally and stay away from proc nuttiness,
> without mucking with ssh and awk.
>
>     grep ^processor /proc/cpuinfo | wc -l
>
> Or:
>
>     ssh -l $TARGETUSER $TARGETHOST "[ -e /proc/cpuinfo ] -a grep
> ^processor | wc -l"

Ooops! I meant


    ssh -l $TARGETUSER $TARGETHOST "[ -e /proc/cpuinfo ] -a grep
^processor /proc/cpuinfo | wc -l"

The check fo rprocinfo is in case /proc isn't mounted.