Randy Calvert

It was what Jim said … I would suggest temporarily removing read/write separation to see if this resolves the issue. The change that made the actual difference removed using separate MySQL read/write shards. You’re only connecting to one database now.

They’ve already said it’s on their radar. It will most likely be like PHP8… released right before the actual EOL. Given there has been no set date (just 2023)… that’s at least 12 more months. and with internet years being like dogs years… there’s still quite a bit of runway left. 🙂

@Miss_B is absolutely right to look at the error log. You mentioned using PHP 7.4, but I don't see any comments about checking to make sure the new host has all of the required PHP modules. Upload that file. If anything is in red, you would want to make sure you address it with your host.

I would just disable everything and have a clean upgrade. Enable them one by one post upgrade.

Are you also a moderator? Admin is permissions in the ACP. Moderators are on the front end. (Add admin group as moderators if you have not already.)

It’s not in the release notes. So I assume not.

It’s most likely all Samsung S21s.

So that the request is not lost in the sea of support requests, you might want to post this ask in the Feature Suggestions forum. 🙂

There is nothing in the default IPB software to do this. You would need something custom that looks at the SSO utilized and places the member into a specific group that has appropriate permissions.

@opentype is spot on. Just make those pages viewable only to the admin group or whatever groups should see it while it's under construction.

Not the lemon!! 😄

If you can't login to the ACP, you need to do what I suggested earlier and try disabling all third party resources using the recovery mode link option I linked earlier. (Check out the 2nd link I posted.) If that does not work, you need to work with your host to pull the error logs to see what is going on. A 500 error just says something went wrong. It does not say exactly what is going on. The server logs should have more detail.

Are these files stored in an existing defined storage method? If they are, migrating them will be easy. You would just provide a ZIP of them when you provide your database. The bigger challenge I see if they’re not in a defined storage method is that attachments and uploads are stored in AWS as https://content.invisioncic.com/SOMEID/filenam. If a different location is hardcoded into posts it’s harder to migrate. It’s not impossible but will take more planning with Olivia, Marc and the migration folks. They would have to run a query across various tables where your attachments are referenced to point to the new location once the migration is complete. If a storage method is designated for those files, it’s super simple to edit the path location of that content. If you do move forward with CiC make sure you call these files out at the start of the conversation so they’re accounted for in the migration plan.

Your server is killing the process because it won't allocate enough memory to the request. You need to have your host increase your PHP memory limit. https://haydenjames.io/understanding-php-memory_limit/

You can also leverage your own AWS S3 bucket and connect to it in any supported method as well.

It will be as good or better than any hosting available. Again with the CiC, you’re not running a site on ONE server but instead UNLIMITED servers. As traffic grows, it automatically scales to handle the traffic. It manages everything for you so you don’t have to worry about the details of server specs or the similar.

A couple of suggestions... Have you confirmed you meet the minimum requirements for IPB? (They've recently changed to require PHP8 for example.) You can confirm using the compatibility checker. There is a good chance if you meet the system requirements for the latest version of IPB, one of your third party plugins/applications is not. You can enable recovery mode to disable them all and restore access to the site.

Yes... but remember... for IPS, there is not a concept of a "hardware crash". IPS is deployed within AWS and uses autoscaling to grow its instances automatically. They also use load balancers to send traffic to multiple servers concurrently. This helps prevent situations like you described where you would have to restore a backup because of a hardware failure. Databases are deployed within AWS Aurora. (You can read more about how Aurora works here.) This depends on what type of data you're talking about. Media files are essentially written to different areas concurrently, one of which is a "permanent state" version. This means that second copy can never be overwritten and is great for continuous backups of static files (things that you upload, themes, etc). Invision has multiple types of database backups... ones designed for "disaster recovery" scenarios should there be a major issue in Aurora or the platform itself, and another designed around the concept of "oopsie... I messed up". Backups are stored within multiple AWS availability zones.

In its original version it was permanently displayed in the ACP. IPB made it dismissible for a period of time before it appears again. Feel free to suggest a change in the Feedback forum. The staff won’t be able to give you much more of a response outside of that. With that said, this is a horse that has been beaten to death previously. As a result, I would not expect much in terms of additional change.

We use the standard. The ones on iOS and Mac look much nicer.

@newbie LAC has not been online for nearly 2 years. The last messages we've seen from him is that he was having health issues. Unfortunately there has not been any word since then. You should most likely consider his resources abandoned and find a different resource.

There have been two different issues in this thread... Related to IPS' CiC platform. This was resolved. An issue related to having mySQL split into separate read/write shards. If your mySQL is not split, it's not the same issue. I would recommend starting a new topic. If downgrading PHP worked, it's most likely a different issue. I would first start by making sure your version of PHP8 meets IPS' system requirements: I would also disable all third party applications/plugins and also switch back to a default theme. In a majority of cases where folks have had issues with PHP8, it's been one of those issues.

It's the same exact software... except as Jim noted... there are a few "extra" things that only work in IPS' custom setup: Available now... Real-time Who's Viewing/Who's Typing Trending module AI image tagging/moderation Coming soon... Live Topics (live chat during a webinar allowing interaction with members during live events that later are turned into forum topic) Video encoding (transform videos into one format that can be played across all different platforms/devices) I agree in that the biggest value comes from not having to deal with all of the underlying plumbing. No worries about installing the software, or making sure you have compatible modules, or finger pointing between host and IPS, etc. Instead of worrying about how to "stitch it all together" between your hosting company and IPS... worrying about things like backup, security, etc... you're giving the full headache to IPS to worry about. This is great for some people to have a solution like Gmail where they just use the software. For others, they prefer to get their hands dirty and control every step in the stack. In those cases, the self-hosted version is a better solution for them.

One of the early lessons I learned... There are three qualities for any technology/company. However you only get to pick 2. Good, Fast, and Cheap....

Take a look at: