How would I go about debugging a cron job that executes the script, but the script seem to not complete?

I have a cron job scrape.sh that look like this:

#!/bin/bash
touch rage
cd /etc/myproject/scraper
scrapy crawl foosite --set FEED_URI=../feeds/foosite.xml --set FEED_FORMAT=xml
scrapy crawl barsite --set FEED_URI=../feeds/barsite.xml --set FEED_FORMAT=xml

When it executes the file rage does get created and judging from my syslog it does run as root, so permissions shouldn't be a problem.

May  6 17:35:01 server CRON[10233]: (root) CM开发者_运维知识库D (/etc/myproject/scraper/scrape.sh)
May  6 17:40:01 server CRON[17804]: (root) CMD (/etc/myproject/scraper/scrape.sh)

When I run scrape.sh it executes as expected and puts the foosite.xml file in the ../feeds directory, the directory exist and is empty when the cron jobs starts. What can I do to solve this issue?

---

1. If I were going to guess the problem it was an environment issue (e.g. scrapy is not in the path).
2. To debug, make sure your cron job is sending the standard out and standard error to a log file/and or syslog

---

Maybe the command scrapy is not found? Cron jobs typically get a different shell environment than interactive shells, so perhaps scrapy is missing in your PATH and you should use /some/full/path/to/scrapy.

If that doesn't help, try redirecting stdout and stderr to some files, so you can see what the output is?

http://tldp.org/HOWTO/Bash-Prog-Intro-HOWTO-3.html