How do I determine if an command completed successfully when run from a
bash script to decide what the script should do next?

Working on a mysqldump backup script which works pretty nice... but
doesn't really care whether the database dump dies, errors out or
completes successfully.

I figure that's kind of important for a backup script.

--
"Remain calm, we're here to protect you!"

On Wed, 02 Apr 2008 16:19:19 -0500, Ivan Marsh wrote:
>
> How do I determine if an command completed successfully when run from a
> bash script to decide what the script should do next?

Depends on the application.
Usually return code is in $?

Example:
ping -c 1 -w 3 $_target_host > /dev/null
if [ $? -eq 0 ] ; then
echo "Ping failure on $_target_host"
exit 1
fi

On some apps I have had to redirect error output to a file
and use grep to check for error string.

On Wed, 2 Apr 2008 21:30:42 +0000 (UTC), Bit Twister wrote:
> On Wed, 02 Apr 2008 16:19:19 -0500, Ivan Marsh wrote:
>>
>> How do I determine if an command completed successfully when run from a
>> bash script to decide what the script should do next?
>
> Depends on the application.
> Usually return code is in $?
>
> Example:
> ping -c 1 -w 3 $_target_host > /dev/null
> if [ $? -eq 0 ] ; then

Oops. that should read
if [ $? -ne 0 ] ; then
echo "Ping failure on $_target_host"
exit 1
fi

For extra points read
man test
http://tldp.org/LDP/abs/html/index.html

On Wed, 02 Apr 2008 21:30:42 +0000, Bit Twister wrote:

> On Wed, 02 Apr 2008 16:19:19 -0500, Ivan Marsh wrote:
>>
>> How do I determine if an command completed successfully when run from a
>> bash script to decide what the script should do next?
>
> Depends on the application.
> Usually return code is in $?
>
> Example:
> ping -c 1 -w 3 $_target_host > /dev/null if [ $? -eq 0 ] ; then
> echo "Ping failure on $_target_host" exit 1
> fi
>
> On some apps I have had to redirect error output to a file and use grep
> to check for error string.

Grepping through an output file is how I have been doing it... which is
prone to not being able to plan for the unknown.

I figured there had to be a more standardized way of doing it.

So I guess I'd have to test the application to see if it produces the $?
and adjust accordingly.

--
"Remain calm, we're here to protect you!"

Ivan Marsh <annoyed@you.now> writes:


>How do I determine if an command completed successfully when run from a
>bash script to decide what the script should do next?

>Working on a mysqldump backup script which works pretty nice... but
>doesn't really care whether the database dump dies, errors out or
>completes successfully.

$?
is the return code of the last command run.

command1 && command2
executes command2 only if command1 completes suddessfully

if command; then
command1

On Wed, 02 Apr 2008 16:43:08 -0500, Ivan Marsh wrote:
>
> Grepping through an output file is how I have been doing it... which is
> prone to not being able to plan for the unknown.
>
> I figured there had to be a more standardized way of doing it.
>
> So I guess I'd have to test the application to see if it produces the $?
> and adjust accordingly.

Just repeating the warning, I have had a good return code in $? and a
failure in standard error. :(
It just depends on the application.

Bit Twister writes:
> Just repeating the warning, I have had a good return code in $? and a
> failure in standard error.

That's a bug.
--
John Hasler
john@dhh.gt.org
Dancing Horse Hill
Elmwood, WI USA

John Hasler <john@dhh.gt.org>:
> Bit Twister writes:
> > Just repeating the warning, I have had a good return code in $? and a
> > failure in standard error.
>
> That's a bug.

Or he's mis-read the manpage's description of return values.

Not everything says "$? == 0" on success. Which process in a chain
does he want to consider the definitive result?

foo > blah | huggghh! ...

He might fiddle with wrapping parts in parens (run in subshells).


--
Any technology distinguishable from magic is insufficiently advanced.
(*) http://blinkynet.net/comp/uip5.html Linux Counter #80292
- - http://www.faqs.org/rfcs/rfc1855.html Please, don't Cc: me.

On 2008-04-02, Ivan Marsh wrote:
>
> How do I determine if an command completed successfully when run from a
> bash script to decide what the script should do next?
>
> Working on a mysqldump backup script which works pretty nice... but
> doesn't really care whether the database dump dies, errors out or
> completes successfully.
>
> I figure that's kind of important for a backup script.

if command args ## replace with your command
then
echo success
else
echo failure
fi

Or:

command args
if [ $0 -eq 0 ]
then
echo success
else
echo failure
fi



--
Chris F.A. Johnson, author | <http://cfaj.freeshell.org>
Shell Scripting Recipes: | My code in this post, if any,
A Problem-Solution Approach | is released under the
2005, Apress | GNU General Public Licence

Chris F.A. Johnson wrote:

> On 2008-04-02, Ivan Marsh wrote:
>>
>> How do I determine if an command completed successfully when run from a
>> bash script to decide what the script should do next?
>>

Normally you get $? as the status for the last command. Most programs uses 0
as failure and 1 as success. See man bash on status is set in different
situations like pipes.

Benny Nielsen wrote:

> Chris F.A. Johnson wrote:
>
>> On 2008-04-02, Ivan Marsh wrote:
>>>
>>> How do I determine if an command completed successfully when run from a
>>> bash script to decide what the script should do next?
>>>
>
> Normally you get $? as the status for the last command. Most programs uses
> 0 as failure and 1 as success.

Nope. Most programs use a return code of 0 to indicate success, and a
non-zero value to indicate the opposite (not always failure).

> See man bash on status is set in different
> situations like pipes.

--
Lew Pitcher

Master Codewright & JOAT-in-training | Registered Linux User #112576
http://pitcher.digitalfreehold.ca/ | GPG public key available by request
---------- Slackware - Because I know what I'm doing. ------