Stuart Silvester

We still rely on an upstream library that makes use of this API. As noted here, we're moving towards native JS in v5.

Hi,

Both of these emails were replied to by our support staff within 30 minutes of them being received. We haven't received any bounce-backs for them yet, so you may want to check your junk folder.

3 hours ago, Square Wheels said:

Is it possible to get a download of a file?

I know IPS took the stance that apps could only be installed / updated through the Marketplace and not downloaded.

I have an app from a developer who has not been around since May 2023.  He seems to have abandoned his site, so there is no easy way to get the app updated.

Would IPS be willing to provide downloads of apps we purchased in cases like this so we can hire another developer to update them?

Thank you

Hi,

No, we're not going to distribute other developers IP through a method they did not agree to.

If you have the app installed, you already have the files on your server. A competent developer would know where to look to get those and build whatever changes you want on top of that.

We have released a patch for this issue.

To apply, please go to AdminCP > Support > 1st (or top left) box will have a "Optional Patch Available" link. Click it and follow the instructions.

We have released a patch for this issue.

To apply, please go to AdminCP > Support > 1st (or top left) box will have a "Optional Patch Available" link. Click it and follow the instructions.

Yes, customers have used the query cache before however it was deprecated in MySQL 5.7.20 and does not exist in MySQL 8 (which we now recommend).

I doubt that would make an impact on TTFB though. It might be worth seeing if you've got any slow queries specific to your community.

This was caused by the RECOVERY_MODE constant being set incorrectly. Removed that from constants.php and everything is working 🙂 

10 hours ago, Cyberg Studio AS said:

...and where exactly do I rename the pages app? in applications\cms\modules\admin folder?

Sorry, that was auto correct. Disable and re-enable the Pages app on the applications page 

It's sounds like you've somehow deleted the hooks.php file. Disable and rename the pages app to regenerate it 

Give the beta a try, see if the new way we handle the icons flushes out the cache. It's unlikely we'll be making any other changes to the beta before final release.

PHP 8.2 will be fully supported in Invision Community 5.

Whilst it may work with earlier versions such as 4.7, it isn't officially supported (there may be issues) and will likely mean that developers mode is unusable..

14 minutes ago, Clover13 said:

I am deleting the CDN files (S3) so it forces an entire rebuild the next time I upload images via the IPS AdminCP and purging the entire site cache in CF.  When I visit the physical links in no-cache private/incognito browser window, they do appear correctly.  Part of me wonders if the Apple device is just not interpreting it correctly OR if there is something specific about each image required in the IPS AdminCP for setting up the PWA (i.e. certain exact sizing is required, or certain image properties are required, etc).  

Right, but if your Cloudfront/Cloudflare configuration said "Cache this static image for 1 year" for example; The device (iPhone) itself is going to cache that resource for a long time and not check the CDN/Website for a newer version.

15 minutes ago, JSensation said:

Hello, thank you for the replay. if the external images where to be https: would that make any difference? Is there any other action I can take other than going in manually?

If they were https:// they'd be displayed inline as expected, however that's something you would need to do prior to the conversion. At this point, fixing them manually would be your only solution.

If you're using a CDN with default cache-control headers for static files, it's very likely that the device itself is caching these resources. The good news is that the current beta (4.7.15) introduces cache-busting for web app icons, so that may resolve your issue after upgrading.

Hi,

By default Invision Community won't embed http:// images (note these are not vBulletin attachments but externally linked images) since they'll cause the page to show insecure warnings. In this situation the image is replaced with a link.

If you're using PHP mail, do you have it specified in AdminCP > Email Settings > Additional Parameters?

That might just be a side-effect of the actual issue (this is where it's trying to report the exception to us), but check that your server has `curl_multi_exec()` available and that it isn't listed in the `disabled_functions` section of the PHP configuration.

A blank page is typically a PHP error. Have a look in your servers PHP Error log for any errors that occurred around the time you tried to upgrade

7 minutes ago, Jon Matcho said:

Thanks @Stuart Silvester, but that was known and not a complete solution to my actual question regardless. @Randy Calvert deserves the credit for the solution in this case.

Regardless, thank you both -- I just wanted to know how the built-in Members group was handled, and the thread referenced above clears all that up.

Sure, the method I mentioned can be set to require moderation for the first two comments/content and then they do not require further moderation. 👍🏻

5 minutes ago, CheersnGears said:

I am stuck in this scenario now myself.  Got into the AdminCP template issue situation, followed the instructions here to uploaded files and upgrade (went to the latest beta). The template issue remains, but now I don't have a version number I can upgrade to.

Have a look in the `core_log` table, order it by newest and see if there's anything logged for the AdminCP URL. You should see something indicating what the issue is.

There's a group setting for this, edit your 'Members' group, go to the 'Content' tab, enable "Require approval before content shows" and configure the limits as required.

In Pages, each custom field has it's own permissions. You'll need to configure these for the title and content field (and any others) to enable guests to see them on the new record form.

7 minutes ago, My Sharona said:

The issue seems to have passed (I was able to faeces post 8 consecutive times in under a minute without getting the puzzle prompt) for myself. I am however getting reports from others that they are still experiencing this matter.

As far as I know, it should stop prompting you have you successfully complete the puzzle.

It's something we're still working on, it may work for some attempts but not others. If it doesn't work, you'll see the message asking you to try again later.

I have passed this feedback to our Cloud team, they'll be reviewing and making adjustments as needed. Normal usage shouldn't be causing the puzzles to be prompted, so we'll likely need to make some adjustments.