[Pig-user] Set visible name of a running pig job

At the beginning of your script, use:

SET job.name 'this is my alternative name';

You can also use parameters like $PARAM in the name.

EG

This will work but will (in .80 at least) change only the part of the
job name that's not 'PigLatin:'.

That is, if you use job.name 'hello' in a script named test.pig
You end up with a full name of:

PigLatin:hello

Instead of PigLatin:test.pig

Just FYI.

Thanks Eric and Mark.

Now I see that job.name is documented in http://pig.apache.org/docs/r0.8.1/piglatin_ref2.html#set (duh). It also says there "All Pig and Hadoop properties can be set."

Trying to figure out what exactly those properties are (is there a list someplace?) I looked at my job configuration at

http://myserver:50030/jobconf.jsp?jobid=job_2011999999_9999

where I now see 'mapred.job.name' --> PigLatin:hello

after I set job.name to 'hello'.

But
SET mapred.job.name hello
doesn't seem to have any effect at all.

So I am confused about which properties I can set, and how to refer to them. Is there a doc or wiki page someplace that explains this to a pig/hadoop novice?

Thanks again for your help.

William F Dowling

-----Original Message-----
From: Mark Laczin
Sent: Thursday, May 26, 2011 2:23 PM
To: user@pig.apache.org
Subject: Re: Set visible name of a running pig job

This will work but will (in .80 at least) change only the part of the
job name that's not 'PigLatin:'.

That is, if you use job.name 'hello' in a script named test.pig
You end up with a full name of:

PigLatin:hello

Instead of PigLatin:test.pig

Just FYI.

Another option is to use -Dmapred.job.name=whatever on the command line.

2011/5/26 <william.dowling@thomsonreuters.com>

Thanks Jonathan. I've seen other references to using -D... on the command line, but I haven't had success with it. I tried
pig -param a=b -Dmapred.job.name=whatever myscript.pig
and the script failed and I got a usage message

Apache Pig version 0.8.1 (r1094835)
compiled Apr 18 2011, 19:26:53

USAGE: Pig [options] [-] : Run interactively in grunt shell.
Pig [options] -e[xecute] cmd [cmd ...] : Run cmd(s).
Pig [options] [-f[ile]] file : Run cmds found in file.
options include:
-4, -log4jconf - Log4j configuration file, overrides log conf
[...]

The usage message doesn't mention -D.

I think there's a bug in command line processing: when I reversed the order of the -param and -D, then I did not get the usage message, and my script ran.

But, mapred.job.name did not get passed through to hadoop. The configuration reported by the jobtracker shows the default name for the job, PigLatin:myscript.pig, not the string following the -D.

So that looks like a different bug. I have not tried the -propertyFile switch. What would be the format of entries in that file:
a b
a = b
<xml>???</>

Thanks again,
Will



-----Original Message-----
From: Jonathan Coveney
Sent: Thursday, May 26, 2011 4:47 PM
To: user@pig.apache.org
Subject: Re: Set visible name of a running pig job

Another option is to use -Dmapred.job.name=whatever on the command line.

2011/5/26 <william.dowling@thomsonreuters.com>