Hi there, V2 has been rolled out and has improved the reliability for the majority of spaces - but we acknowledge that you still may be seeing this issue for larger spaces above 2000 pages. The team is investigating paths forward to address this. Please follow https://jira.atlassian.com/browse/CONFCLOUD-75912 for additional updates.

I'm stuck at 85% trying to do a very unexceptional PDF export on Jira cloud. At time of writing I'm up to 3100 seconds. A minimal set of selected pages, maybe 20 pages total, no attachments, no images, just text and lists with a few headings and links. I even removed the one panel macro that was in there and some Jira issue macros as I thought that might be creating the problem, but it appears that's not it. There's no message indicating what's wrong and I don't know where to look for one.

In case it's of any use, when I open the Chrome console I can see a 409 error on https://.../messages/confluence-external-collab-user-in-space/start

We are still unable to proceed past 50%? Any other updates?

@Alma Younger, we are still getting this issue in our cloud instance - is this 100% rolled out to all cloud instances?

I am closing this bug as we are currently rolling out page and space PDF export V2 which should resolve this issue. It is rolled out to 25% as of today and is expected to be rolled out to 100% on May 8th. 

Hi everyone,

To follow up on sbao's comment above, we are close to wrapping up our development work on this feature. However, we are intentionally planning a slower roll out based on our previous learnings. The date on this ticket has been moved accordingly.

Hi everyone,

Thank you for your patience as we work on fixing this bug. We’re currently making some larger architectural changes to PDF Export that should resolve this bug (and many other PDF export related issues) that started rolling out in Q3 2022 but needed to be rolled back in early November. The team is currently working to re-roll out the improvements by mid EOQ1 2023.

We know you have been waiting eagerly for this fix as we are, and we will be providing any updates as they come. Thank you again for your patience.

Best,

Steven

Doing the export in several parts takes time on our production (export time and time to find the problematic pages). Export is possible through a third-party plugin, but the trial version is only 30 days.
This feature does not pose any problem on the on-premise version. For the moment we only have the impression of a downgrade with the cloud version, the development does not seem to be faster despite the fact that everyone has been forced to migrate.

We're also held up by this issue.  The space that's failing in this exact fashion does not have 1000 or more images, though.  Are others seeing the failure only with 1000 image use case? 

Unfortunately, I don't think I can.  Only site administrators can raise tickets, and I'm not a site admin

Hi 75b7b0fec79b,
Please raise a ticket on support.atlassian.com. With the site details on the case, we should be able to confirm and check if there is any issue.

Regards
Atlassian Cloud Support

Is there a way of determining whether the fix has been rolled out to my space?   I'm still getting the 50% error.  Somewhere I read a suggestion that it can't cope with restricted pages and to deselect these.  I've tried that (assuming I've correctly remembered which pages are restricted), but I still can't get the full export to work?

We have started the rollout of the re-architected space-level PDF Export functionality. As communicated in this post (https://community.developer.atlassian.com/t/improved-confluence-page-pdf-export-only-backend/60637) re-architected page-level exports have already been released to 100% of customers. Now we have started rolling out the same exported PDF experience for space-level exports as well. We expect the roll-out to be at 100% by end of November '22.

Thank you for your patience while we were hard at work to bring high-quality, wysiwyg PDFs to you.

Hi there,

Any updates on this bug? A small improvement that could help the quality of life of the PDF export function is to save the custom checkbox preferences from last time, if the export fails. At least then we can tweak our export until it's small enough to pass HTML validation.

All the best,

Hector

This issues is still happening. Any workaround / patches?

This bug still happening on the cloud version, so not even access to the temp folder anymore.

As mentioned in previous comments, this is not minor as we need to share documentation with external parties that don't have access to our intranet.

This bug is still happening. What is the solution?

Any progress to report? Early 2022, has come and gone - we're now into the second quarter. 

Hi everyone,

Thank you for your patience as we work on fixing this bug. We’re currently making some larger architectural changes to PDF Export that should resolve this bug (and many other PDF export related issues) that is set to fully release in early 2022. We acknowledge that this has run into delays as the development team wants to ensure that we will not regress on functionality with the architectural improvements.

We know you have been waiting eagerly for this fix as we are, and we will be providing any updates as they come. Thank you again for your patience.

Best,

Steven

I bumped in the same issue with 551 pages.  For an escrow I need to export one space with all pages inside of it. I switched from PDF to HTML export, but I did not see any progress on HTML export, just the time counter was increased and the progress bar stuck with 0%. If I tried to export a subset of HTML pages, Confluence tells me that my export is still ongoing, but I do not see any further details or location for intermediate export results. It is very common for us to create an escrow at the end of each project. 

Hi everyone,

As we work on the larger architectural changes we are trying to see if we can solve this issue for you faster with a few options we're weighing. 

It would greatly help if you chimed in on the necessity of the following features for space PDF export - 

1. Table of Contents w/clickable links
2. Page Numbers
3. Custom Stylesheets 
4. Headers & Footers

Furthermore, does the current behavior below of PDF space export match what you expect?

1. Header Demotion (Headers are demoted depending on each page’s depth in the page hierarchy. e.g. Header 1 in the top page will be header 1, but Header 1 in a page 1 depth below will be converted into Header 2)

1. Page Concatenation (each Confluence page is not started on a new page, but instead starts right after the previous page ended)

Best,

Steven

Hi everyone,

Thank you for your patience as we work on fixing this bug. We’re currently making some larger architectural changes to PDF Export that should resolve this bug (and many other PDF export related issues) that is set to fully release by the end of October.

We know you have been waiting eagerly for this fix as we are, and we will be providing any updates as they come. Thank you again for your patience.

Best,

Steven

Unbelievable that no work has been logged in over a year. 

I see this has been categorises as minor - considering Confluence is a documentation tool not being able to export documentation is not minor. We need to provide documentation to third parties who do not have access to our confluence system. How are we meant to achieve this if the export does not work?