Issue with non-open source PDFs with proxy

Report ID: D506385811

I am not able to save PDFs that are not open-source when connected my university's proxy.

Steps to reproduce:
1. Access site https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.132.186001 on google chrome
2. Use my university's EZProxy to reload the page with university access
3. Accept zotero banner to redirect future requests to ezproxy
4. Hit zotero plugin
5. Red X appears in front of "Full Text PDF"

Proxy settings:
Login URL scheme: http://ezproxy.princeton.edu/login?qurl=%u
Proxied URL scheme: %h.ezproxy.princeton.edu/%p
Hostnames: journals.aps.org
  • edited May 4, 2024
    Can you provide a Debug ID from Zotero (rather than the Zotero Connector) for reloading the page and trying to save unsuccessfully?
  • D1489474532

    Thanks!
  • Hi I'd like to reopen this issue. Please let me know if I can provide more debugging information.
  • Is this happening on all sites or just that one?

    Also, it looks like Zotero is connecting via a system-configured proxy server (and it should be prompting you for the password for that at startup). It's a bit unusual to be using both an actual proxy server and web-based proxy, and it's possible they could be conflicting — e.g., if your browser isn't actually using that proxy server, then the file-download request (which is made by the Zotero app itself) would be coming from a different IP address. So to start, you'd want to make sure that both Zotero and your browser are connecting to the internet in the same way.

    Beyond that, it's hard to say what might be happening, but you should try disabling your browser extensions and reloading the page and/or trying from a different browser.

    We're working on some changes to how Zotero downloads files from the browser that should help avoid situations where files are accessible in the browser but fail from Zotero.
  • Thanks for taking another look.

    It's happening at all sites that don't have open access. Here's another example:
    https://pubs.aip.org/aip/apl/article/124/21/212101/3294292/Amorphous-TeO2-as-p-type-oxide-semiconductor-for?searchresult=1

    I still run into issues when accessing from a different browser, even when I'm on my school's VPN so it should not be using the proxy. I'm not sure why the app and the connector are connected to the internet in different ways... I typically just connect to my home wifi. I also am not prompted for a password at the startup of the app like you suggested.
  • PAC file installed from http://127.0.0.1:9000/systemproxy-f0174d73.pac

    (3)(+0000053): Proxy required for https://www.sciencedirect.com -- making HEAD request to trigger auth prompt
    There's a PAC (Proxy Auto-Configuration) file set up on your computer that's instructing apps to use the proxy for academic sites. It's possible this is related to your VPN setup, if the VPN is configured to only be used for certain sites. This also may or may not be causing the failure. In any case, this is something you would have to debug.

    (And sorry, I was wrong when I said it would necessarily prompt for a password.)
  • Thanks for your help. I was able to figure out the issue and wanted to share in case someone has a similar issue in the future. My work has zscaler installed on my computer which is an internet security application and interferes with the zotero proxy. When I disable zscaler, there are no issues retrieving pdfs.

    Thank you
  • Yes, Zscaler is known to block Zotero 6, because it identifies as an older version of Firefox. The Zotero 7 beta will work.
Sign In or Register to comment.