Our 2nd-tier support team has identified a quick workaround: clone the page using the Yoast Duplicate Post plugin; the cloned page doesn't have the issue.
This may be a good workaround for now.
Our 2nd-tier team is still working on another fix for the issue, and I'll update you as soon as I receive their response.
Our 2nd-tier support team has found that it's a database issue. The workaround is to use the Yoast Duplicate Post plugin, as the duplicated posts won't have the same issue.
Please let me know if Hetsher has tried the workaround.
I understand the suggested workaround using Yoast Duplicate Post, and I appreciate the confirmation that this is indeed a database-level issue.
Before applying this workaround, I want to be careful, as this site has:
existing indexed URLs,
multiple languages,
and SEO-sensitive content.
Cloning pages would create new post IDs and URLs, which is not ideal as a structural solution and may introduce additional SEO side effects.
Given that 2nd-tier support has identified this as a database issue, I would prefer to address it at the source if possible - for example by:
cleaning or rebuilding the affected WPML translation records,
resetting locked translation jobs,
or repairing inconsistent entries in the WPML tables.
Could you please confirm whether a database-level fix or cleanup is possible on your side, now that the root cause has been identified?
If needed, I can of course test the workaround on a single non-critical page, but I’d prefer to avoid duplicating live content unless there is no alternative.
1) Please try the following on a non-critical page:
1- Edit the page in all languages and change the slug to slug-old
2- Duplicate the page in the default language
3- Change the slug from slug-old-copy to slug
4- Check the issue
2) Our 2nd-tier support was not able to find an alternative workaround, but I'll share your concerns with them.
I’ve completed the test you requested on a non-critical page. Below is a clear and final summary of the results.
Result of the test
After duplicating the page and assigning the original slug to the duplicate (test performed on the Privacy Statement page, Dutch → English):
The WPML translation editor opens normally
The translation can be saved / closed without redirects or warnings
The editor no longer falls back to the WordPress/Elementor editor
This confirms that the original blocking issue was caused by legacy database relations tied to the original post ID, not by the page content itself.
Remaining issue (important clarification)
Before these problems started, my site worked for years in local translator mode:
I only saw pencil icons
I edited translations directly myself
After saving, the icon returned to a normal pencil
Translations were never treated as “jobs” or reassigned
After the workaround:
The editor is functional again (good)
However, WPML now treats the translation as a translation job:
gear icon instead of pencil
messages such as “waiting for available translator”
in WPML → Translations, the same page appears with conflicting statuses (e.g. “needs translation” and “completed but needs update”)
This workflow behavior did not exist on my site before the issue and indicates remaining inconsistencies in WPML’s translation records.
Conclusion
The workaround successfully restores the editor and proves this is a database-level issue.
However, duplicating pages one by one is not a sustainable solution for a multilingual, SEO-sensitive site, and the original local-translator workflow has not been fully restored.
Question
Now that the root cause is confirmed, is it possible for 2nd-tier support to perform a database-level cleanup/reset of WPML translation data so the site can fully return to the original local-translator workflow (pencil icons, no translation jobs), without duplicating pages individually?
Thanks again for your help and for escalating this to 2nd-tier support.
I am following up on the issue that has been blocking my live WooCommerce website since December 8.
I understand that this case has been escalated to 2nd-tier support. I also acknowledge that a workaround was suggested (duplicating pages). However, this does not resolve the underlying problem. Pages remain marked as “outsourced to another translator / in progress”, which prevents me from:
editing content,
managing translations myself,
and safely updating or adding products.
As a single site owner and editor, this effectively means I cannot work on my website.
At this point, I need one of the following:
A workable temporary solution that restores full editing control, or
A clear and realistic timeline for a permanent fix.
Could you please clarify:
whether this issue has been identified internally as a known bug,
whether a developer is currently assigned,
what exact action is being taken,
and what the next concrete step will be?
I would appreciate a clear update on how and when this will be resolved.
Sorry for the late reply. Please downgrade WPML CMS 4.7.6 and String Translation 3.3.0, update an Elementor page, and check whether the issue persists.
You can find the previous versions for WPML CMS here and ST here.
Please take a full backup of your site before running the test above.
Before proceeding with a downgrade, I’d like to clarify one point.
Based on the previous test, we’ve already confirmed that:
the editor issue is resolved after duplicating the page
the problem is tied to legacy translation/workflow records, not the content
the remaining issue is that WPML still treats translations as jobs instead of restoring the original local-translator workflow
Given this, could you please clarify what exactly the downgrade test is intended to confirm?
Specifically:
is the goal to identify a regression in recent WPML versions, or
is it expected to clean up existing translation/workflow data?
Since this is a live site using Elementor and WooCommerce, downgrading WPML and String Translation carries some risk, even with a backup. I’d therefore like to be sure this test is meaningful for the database/workflow issue already identified.
If 2nd-tier support believes a database-level cleanup or reset is still required, I’d prefer to focus on that path instead of a rollback test.
Thanks for your understanding, and I’m happy to proceed once this is clear.
1) Please try this test on your staging site, as the issue persists there as well.
2) They believe this behavior may be related to the media handling changes introduced in WPML 4.8. This conclusion is based on call stacks captured on the site, which show media translation processes being involved when the translation status is reset from “needs update” back to “up to date.”
To verify this, they temporarily rolled the local site back to WPML 4.7.6 and String Translation 3.3.0. Under these versions, the issue no longer occurred, and the translation status was updated correctly when the original content was modified.
Thanks for your cooperation
Best regards,
Osama
The topic ‘[Closed] WPML Translation Dashboard stuck on “Loading items…” – possible ATE/queue issue’ is closed to new replies.