Skip to main content
Participant
September 24, 2019
Question

PDFs corrupt after network disconnect

  • September 24, 2019
  • 3 replies
  • 2946 views

Hi everyone,

 

in our (large) company, we use Adobe Acrobat Standard/Pro 2017 for PDFs. We have thousands of PDFs stored on a server. 

 

When opening the PDFs from the server, they work as expected and we then keep them open in the background. However, as soon as we loose connection to the server only for a couple of seconds (happens very regularly, i.e. a) because of sending computer to standby, b) undocking computer from docking station so connection changes from ETHERNET->WLAN, c) a bad WLAN connection including a change of WLAN network while walking through a building), all the PDFs that were open before cannot be viewed anymore. This issue remains even when the WLAN/network connection is fully re-established! They either have empty pages, or an error like "There was an error reading from the stream" or "There was an error reading this document (109)" occurs. Given that this issue appears on Adobe Acrobat & Adobe Acrobat Reader, but NOT in the Chrome-internal PDF plugin, it seems like Adobe is not "properly" caching the PDF files. In company-life, this is extremely frustrating as we regularly have to reopen the ~6 PDFs that we work with concurrently.

 

Is there a fix or setting available from Adobe? This and related issue seems to be known and plague many users on the internet, see https://community.adobe.com/t5/Acrobat-Reader/After-some-hours-pages-go-blank-in-Acrobat-Reader-DC-and-it-says/td-p/10148552.

 

Note: Happens on both Win7 and Win10, and affects basically every colleague in my team who uses Adobe Acrobat / Reader.

 

Best

Philipp

    This topic has been closed for replies.

    3 replies

    Participant
    November 15, 2022

    My coworker was opening .pdfs this morning. Then all of a sudden was unable to open the same .pdfs with an opening error. Found out, she had undocked, went to a conference room and connected to the corporate wifi.  After reading this post, we had her clear her cache. She is now able to open the .pdfs that she was getting an error on.  So Thank you!!!!

    philippoeAuthor
    Participant
    September 27, 2019

    FYI, I now posted a bug report here https://acrobat.uservoice.com/forums/590923-acrobat-for-windows-and-mac/suggestions/34809364-my-adobe-acrobat-pro-dc-constantly-looses-page-con . The issue is that many different user reports on this issue exist at uservoice.com, but they are very fragmented. Please upvote the respective issues there if you hve this problem such that Adobe is made aware of it.

    Participant
    November 17, 2022

    Did you ever disover a solution to this? I am having the same issue within my organization!

    Legend
    September 24, 2019

    Correct, Adobe software does not cache PDF files. They are read on demand, so that huge files can be handled efficiently. What is read may be held in memory, but memory use is deliberately limited. If the connection closes, it cannot continue to read the file, so parts not in memory will give an error.  I think you have a strong handle on the cause, most people reporting this believe it is random or just something that happens after a few hours.

    philippoeAuthor
    Participant
    September 24, 2019

    Thanks! Is there a setting or registry key to a) store all data in cache all the time or b) just increase the amount of memory that Acrobat reserves for caching? And just to be clear, i am not asking to change the default behavior of Acrobat, just to give an option to change this.

    Legend
    September 24, 2019
    No, there does not seem to be any such option. I wonder if there is some client configuration that could be changed. So far as I can see, Reader opens a file, and reads it on demand, which is a normal and common thing to do. Your network interruptions mean that somehow the file is closed, but the client is not told, presumably. The client keeps trying to read and keeps getting an error. So, the question might be: in my LAN, why does a momentary interruption leave file handles invalid on the client?