[cs615asa] HW#6 Question 1

[Tianxiao Yang]

Hi all,

I got 3 problems:

1.

By default, afewmore prints the instance IDs of the newly created EC2
     instances as the only output.  Unless an error occurs, no other output is
     generated.

This instruction tell us to print IDs of *newly created *ec2 instances, but
in the example specified below:

The following examples illustrate common usage of this tool.

     To create ten more instances of the EC2 instance i-0a1b2c3d4f and copy
     the contents of the '/data' directory from that instance:

	   $ afewmore i-0a1b2c3d4f
	   i-0a1b2c3d4f
	   i-1a1b2c3d4f
	   i-2a1b2c3d4f
	   i-3a1b2c3d4f
	   i-4a1b2c3d4f
	   i-5a1b2c3d4f
	   i-6a1b2c3d4f
	   i-7a1b2c3d4f
	   i-8a1b2c3d4f
	   i-9a1b2c3d4f
	   $ echo $?
	   0
	   $

"$ afewmore i-0a1b2c3d4f" this command generated ten IDs, but one of
them is the original ID itself, there are only 9 newly generated IDs.
Do we need to count the source instance itself as newly generated
instance?

2.

     -d dir   Copy the contents of this data directory from the orignal source
	      instance to all the new instances.  If not specified, defaults
	      to /data.

By default(-d not specified), source data directory should be /data,
but in the example below:

     To create just one more instance and copy the contents of the directory
     '/usr/local/share':

	   $ afewmore -n 1 i-0a1b2c3d4f
	   i-1a1b2c3d4f

	   $

In the command, -d option is not specified, but this command "copy the
contents of the directory '/usr/local/share'". Which one is the right
default directory?

3.
User who uses afewmore cannot chose the directory to which the contents of
source data should be copied.
-- Can we just copy the source data to home directory of newly created
instance?
-- If we need to copy the /data directory from source instance A to the
/data directory in newly created instance B(doesn't have a /data
directory), we usually need to execute the operation as superuser for
instance B. How could we know the what is the superuser command(sudo or yum
etc.)?
-- If /data directory in A is a mount point for a filesystem, do we need to
create the same filesystem in B and mount it to the same directory?

Thanks,
Tianxiao Yang




On Sun, Apr 9, 2017 at 10:54 AM, Jan Schaumann <jschauma at stevens.edu> wrote:

> Josh Dolisca <jdolisca at stevens.edu> wrote:
>
> > I was wondering if there were any restrictions on how the command
> > should be run. Does it have to be packaged to the point where it is a
> > standalone command? For example, if we have the tool written in perl,
> > can we include in the README that the tool should be run as "#
> > afewmore.pl<http://afewmore.pl/> [...]", or does it have to run as "#
> > afewmore [...]" from the command line?
>
> Your tool should be able to be run as "./afewmore".
>
> -Jan
> _______________________________________________
> cs615asa mailing list
> cs615asa at lists.stevens.edu
> https://lists.stevens.edu/mailman/listinfo/cs615asa
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.stevens.edu/pipermail/cs615asa/attachments/20170409/87679207/attachment.html>