Skip to main content

modsec denies Amazon crawler access to JPG images with 403

Comments

2 comments

  • Nick Fenwick

    I found that you can disable that rule by its id 920600 globally with a single line in the apache conf:

    SecRuleRemoveById 920600

    I have small config files in /etc/apache2/conf.d/userdata/ssl/2_4/<username>/<domainname> and can use the following three commands to rebuild the apache config and restart the server:

    # /scripts/rebuildhttpdconf
    # /scripts/restartsrv_apache
    # /scripts/restartsrv_apache_php_fpm

    It would be nice to use a more precise rule to only exclude the Amazon crawler, but it doesn't seem to use a sensible user agent or other string I can identify it by ("User-Agent: Java/1.8.0_442", really Amazon?).

    Open to any other suggestions, my inventory import is now succeeding 100% as Amazon can now request my images OK.

    0
  • cPRex Jurassic Moderator

    Hey there!  From the cPanel side of things, whitelisting the rule is the best option I would have available.  Some users have went deeper into the regex of the rule and customized that:

    https://security.stackexchange.com/questions/267800/updating-regular-expressions-for-the-rules-920600-921421-921422-and-922110-in

    but that isn't something we'd support on our side.  You're welcome to try it if you want to go that far, but I can't say if it will cause other issues or not.

    1

Please sign in to leave a comment.