Skip to main content

Apache logging issues after NGINX Manager uninstall

Comments

8 comments

  • aluputi
    After some troubleshooting, I think I've found the issue. The file /etc/apache2/conf.d/includes/pre_virtualhost_global.conf still contains the following, which should be removed: RemoteIPHeader X-Forwarded-For LogFormat "%a %l %u %t \"%r\" %>s %O \"%{Referer}i\" \"%{User-Agent}i\"" combined RemoteIPHeader X-Forwarded-For
    0
  • cPRex Jurassic Moderator
    Good job tracking that down - cPanel doesn't support the Nginx Manager tool, so it likely wouldn't properly remove everything when uninstalled from a cPanel environment.
    0
  • aluputi
    It does, it's integrated in cPanel:
    0
  • cPRex Jurassic Moderator
    Thanks for that - since you said you were uninstalling just a manager, I had assumed it was a different tool, as our Nginx Manager itself can't be uninstalled. On a test machine, I installed Nginx through the Nginx Manager tool, which just installs the package the same as it would through the EasyApache 4 interface. However, I don't see that any of the Apache include files were modified as part of that work. Is it possible there was some other customization done on the machine as well?
    0
  • aluputi
    No, there are no special customisations. From WHM, when accessing the NGINX Manager option, you can also uninstall it. Apache files are modified when installing / uninstalling NGINX Manager. You can clearly what's being changed on install / uninstall, because the interface has a live log that shows everything being done. The code I discovered that isn't removed is required for the correct IP passthrough from NGINX back to Apache, so it's clearly added by the NGINX installation.
    0
  • aluputi
    And first of all, I decided to stop using NGINX Manager, because one day, after upcp cron job ran, something broke. Some NGINX configurations had errors that prevented it from running, making all the hosted websites go offline. This was a few months ago and this happened simultaneously on more servers that had the jobs running at the same time. I think better testing should be done before pushing updates to RELEASE branches... or maybe I should switch to STABLE.
    0
  • cPRex Jurassic Moderator
    Could you submit a ticket with our team so we can check your system? I'm not seeing that code added when I did the installation on a test machine, so I can't say for sure where that would be coming from.
    0
  • aluputi
    Sorry, but that isn't an option. I can't allow any external access to the servers I manage. Maybe that code was added there by the version of NGINX Manager I had installed in the past, I don't know what the current version does.
    0

Please sign in to leave a comment.