-
Bug
-
Resolution: Fixed
-
Highest
-
95
-
Severity 2 - Major
-
125
Editor fails to load
Steps to Reproduce
- Go to a published page in Confluence Cloud
- Click the Edit (pencil) Icon
Expected Results
Editor loads as expected.
Actual Results
Editor fails to load with the following message displayed in the upper-right hand corner of the window:
Something went wrong after loading the editor. Copy your unsaved changes and refresh the page to keep editing
The below error is seen in browser console:
Synchrony: Error: entity destroyed
Workaround
As a workaround, for now, you can refresh the page each time this happens and it should fix the issue.
[CONFCLOUD-65514] Editor fails to load - "Something went wrong after loading the editor." message displayed
Hello together,
I can´t really see a solution for this issue so why is it closed?
I demand for an answer
for our problem, we used workaround to change the Collaborative Editing to Limited in Admin panel. that allowed use to edit the pages.
To resolve our issue, we found out that commons-lang3-3.9.jar file was not at it's default location of [ConfluenceHome]/synchrony-proxy/WEB-INF/lib. Verify if that jar file is there. we moved the file to right location our issue was resolved the issue.
Same problems here. We use it as a server version and have about 800 users.
Can you tell us when there will be the next bugfix?
The publishing software can't publish. And this ticket is closed? Is this going to be fixed ever? Or it is time to switch platforms now?
hello i have this problem on a Confluence cloud installation of about 500 users, when do you plan to solve it forever
Hi,
I contacted the Atlassian support regarding this issue and its frequency, and they told me that it could be related to the migrations from the legacy Confluence editor to the new one. In fact, below is their answer:
New page editor on Confluence creates but does not close connections to Synchrony during single page app transitions when viewing and editing pages. These transitions stack connections over time, consuming the available connection pool in the Synchrony cluster. Monthly active users for the new editor has increased as the Confluence team have progressed their rollout.
Synchrony is the service that runs the collaborative editing features in Confluence. As far as I can tell, the increased load due to the gradual migration of the new editor was unexpected. Our developers are aware of the issues and have mentioned the corrective actions:
- Clean up logging(reduce noise) to make early detection easier
- Increase number of metrics measuring new editor rollout events - helps with root cause analysis
- Verify exponential backoff functionality in the client works as expected - should be already be working
- Enhancement and further review of the capacity planning around the migration
On my side, I sometimes continue to experience this issue, and the workaround mentionned in this bug report (refreshing the page each time this happens) still works. But it is very annoying when I realize that not all the modifications made are not taken into account after refreshing.
Could you please open another Bug Report / reopen this one, in order to minimize (or cancel) the effects of migrations to the new editor on our sites?
Thanks in advance.
I am getting a similar issue. Creating a new page using the + button on the navigation bar is fine, however when I edit an existing page and enter a link to a page I want to create new, this is where I encounter an issue. Refreshing does not resolve this problem either, so I have to create using the 'c' or '+' sign. Not a huge problem for me but still a little irritating. I'd prefer incremental feature delivery than lower priority niggles like this. It does stress out inexperienced Confluence users though, but I do my best to help them out.
I agree complete with John Lammers
But dear colleagues from Atlassian:
For the daily work it is much more helpful, to work without these ugly bugs... this causes day by day just useless time without added value
This long-standing bug has still not been addressed.
Atlassian doesn't seem to have the willpower to admit that this bug is not fixed and to dig into figuring out the root cause. After all, working on new features is so much more fun than fixing bugs ... for both developers and product managers.
We've solved this issue yesterday for Confluence Data Center, thanks to the help of Atlassian support.
In our previous configuration we had synchrony running on 2 Linux nodes. But after upgrading to version 6.15.6 the errors occurred.
In this new version synchrony is selfmanaged by Confluence ( see Clustering in the General Configuration). We turned off the external Linux nodes and openend port 25500 on the firewall.
We are also facing this issue more frequently.... Refreshing the page does not solve it at all
Disabling collaborative editing fixes the issue completely, if that is an option for somebody, it was the best option for us
This issue persists for me as well. It's intermittent but quite frustrating and disruptive.
Hi tony.benbrahim1, sorry to hear you are having issues with the latest Server version. I am from the Confluence Server team and I have had a look but can not find any open bug for this issue in the CONFSERVER project so I am unable to link you to any open bug report. If you have not done so already I suggest you reach out to the Confluence support team and they can help you identify the cause of your issue. Details for contacting our support team can be found here https://support.atlassian.com/confluence-server/.
So I have the latest release of server, and still see this. This is going to be hard to explain to users
This bug is not fixed. I've seen it several times this week.
Seriously, who closes a ticket when the bug is still occurring?
See "Scrum Definition of Done": https://www.scruminc.com/definition-of-done/
@All, has someone already implemented Chris Michaels solution? Did it work for Confluence Data server?
We are seeing same error on server version, wondering if there is another ticket for Server?
Chris Michael,
You mentioned the reverse proxy configuration, I guess you are also using Confluence server?
I fixed this issue by editing a block in the reverse proxy configuration to as below:
ProxyPass /synchrony-proxy http://<domain>:8091/synchrony <Location /synchrony-proxy> Require all granted RewriteEngine on RewriteCond %{HTTP:UPGRADE} ^WebSocket$ [NC] RewriteCond %{HTTP:CONNECTION} Upgrade$ [NC] RewriteRule .* ws://<domain>:8091%{REQUEST_URI} [P] </Location>
Mainly I changed following things
"ProxyPass /synchrony" to "ProxyPass /synchrony-proxy"
"<Location /synchrony>" to "<Location /synchrony-proxy>"
The issue persists even after restart of Synchrony or Confluence. Refresh doesn't work either. Using 6.15.2
I have received user reports of this occurring today, so I too do not believe the issue is resolved.
I still get this message more often than not when editing a page.
It doesn't seems to be resolved.
Refreshing the page does not solve the issue for me. It's been occurring for the past several weeks. TIA
Hello, We have this issue to on the Data Center. Is there for also a fix available?
Thnx
@JoAnna : Thanks for your comment - the problem seems to be solved.
I Have performed several tests.
@tobias they pushed a fix yesterday afternoon that helped my situation. Are you still having this problem?
In the meantime, I have to reload approximately every second page two or more times to edit them.
Sometimes it helps to push F5. If this does not help, I have to to interrupt and start the editor once more
@klaus Started to see the error after upgrading to confluence 6.15.1. Reloading the page doesnt help.
I'm going to close this since there hasn't been any new reports in a while and the root cause for this has been fixed. Editor not loading may still happen for variety of reasons, if this happens please contact support and we'll investigate what went wrong.
Seeing this right now. Error in console is "Failed to load resource: the server responded with a status of 500 ()".
No workarounds address this issue.
The workaround doesn't work on our instance. In addition, I'm seeing this error on a large page I just created - @20 pages : Connection timed out.
Found a workaround! Can confirm we're seeing the same bug as many other users. Seems to be an issue re: how you navigate to the page.
Experiencing the same with a knowledge base hosted in confluence and accessed via service desk. Workaround is below.
Steps to reproduce our bug:
- Go to spacename.atlassian.net/projects/spacecode/knowledge to see your index of wiki pages
- Click on a page
- Click on "Edit in confluence"
- The "Something went wrong..." message appears and page is not editable.
Workaround is based on how you navigate to the page:
- Browse to your space for your project: spacename.atlassian.net/wiki/spaces/spacecode/overview
- Click on the page you want to edit from the sidebar under the "Pages" submenu: spacename.atlassian.net/wiki/spaces/spacecode/pages/someID/pagename
- Click on pencil on top-right corner of page
- Page is now editable - editing seems fully functional
We are still seeing this issue which is now affecting a lot of users - can this be re-opened and investigated?
Issue is still happening with my images heavy page as well. Not all the images will load and I get the UNKNOWN_MEDIA_ID is not a valid file id error. Very irritating. Not sure why this issue is marked as resolved?
We are still seeing this and nearly every page we publish/update today isn't saving?
I've seen this today too but it's not every single edit I attempt so I can get around it by refreshing a couple times. Crazy I'm paying for a service I can't use. We'll see how long I keep doing that.
One of our users just encountered this issue moments ago, which is the first time I've heard of it happening to us. I tried to edit her page and confirmed the issue.
We were able to once more edit the page that had been...ornery...before. So the fix appears to have been at least somewhat effective.
This is still happening for me. The workaround is working, but I'm not able to "Edit in Confluence" from a published page.
We have deployed the fix for this. You should no longer see this happening.
I, too, am facing issues with this today. I am trying to edit a restricted access page with a table in the page. I am the site admin and am added with edit privileges to the page.
I've got a page with just a plain table and text that I use to document contact info that I can't edit. I can't agree with the last two comments, from my experience.
To Ginni Stanton's point, there are significant performance issues with Confluence referencing Jira via JQL script macros which have been well documented.
https://jira.atlassian.com/browse/CONFCLOUD-54881
- When Confluence and Jira are used as a single source of truth, it is unacceptable for pages to fail if there are ~40-50 JQL script macros on a page.
- In my experience, I've seen Confluence pages get finicky with 30 JQL script macros.
- The fact that it has been known and unaddressed for nearly 2 years is bonkers (1 year 42 weeks)
It seems to affect pages that have Macro's or lots of information. I don't have the issue much on pages with just Text.
I've seen this for months now but today it got worse. Refreshing is not helping and I can't do anything.
Same problem here since several days, using Confluence Cloud. Please reopen this.