Matt

Only when using a smaller screen where it drops into the menu button on mobile view. There's no way to hide it using desktop (unless you choose to use the classic view where there is no sidebar!)

Ehren will be covering the theme editor in more depth very soon. I stole a little of his thunder with that gif but I see a lot of concern about new views and how they fit in with their community, so I wanted to show just how easy it is to toggle between them. It's also worth mentioning that these layout options are per-theme and not global as they used to be, so you can mix and match across your community as you see fit.

It is indeed an option.

No two sites are the same, which is why we've made it very simple to switch between layout types. More on that very soon.

Ehren is a superstar, he's barely scratched the surface on what he's been working on for the past six months. I'm really excited by these new views and the sidebar mode gives a very desktop app like experience, whereas the new mobile view is truly fit for purpose and not just a squished desktop view.

I'm honestly unsure. I think it's two factors, the first being that the downtime was intermittent with degraded performance (1-2 mins down, then 5-6 up) - but pages were being returned and that it only affected about 15% of our client sites. I guess Uptime rounds up or down and it wasn't statistically significant when balanced with the entire network and 24 hour period. If you look at our site which was in the affected 15%, then you can see the short periods of true downtime between slower page loads - but it was a few minutes at a time.

When we are hit by a network attack (which happens daily), our automation launches new instances to meet demand. On some occasions (a few times this year) there may be some degraded performance for between 1 and 5 minutes while scaling launches. Earlier in the year we had an outage with our upstream provider which was incredibly frustrating. No one likes degraded performance or downtime, let alone our team who are woken by automated alarms at 2am to come and deal with it, but it's the nature of the internet. When you compile a list it may look significant, but it's still just tiny blips in an otherwise robust service. We really do our best to minimise disruption and we'll continue to revise our alarms, scaling and database instances.

Looking through David's ticket history, I can see there have been some difficult to track issues that impact his community only and aren't indicative of the greater experience, these seem more around database optimisation for his community or slow outbound connections. It's something our support team are working with David on.

Hi David, Any period of degraded performance is frustrating for us all. We have put a lot of energy and cost into ensuring our systems stay online and weather even the most persistent online storms but very rarely some do get through and we need to tweak our protection. Looking back at our independently monitored uptime, we have a 99.9% uptime record with very few periods of significant downtime since Jan 1st 2023. What happened in the early hours of the morning was inconsistent with our long standing performance records and we do not take it lightly. We continue to improve our systems and will keep you updated of any changes. Thank you again for your patience. You, or anyone else reading is very welcome to submit a support ticket if they have any questions or concerns. You are welcome to bookmark and review our independent status page: https://status.invisioncommunity.com P.S. I'm going to move this to the existing topic.

We're still gathering data, but we experienced a significant network attack between 11:00pm and 3:15am EDT which affected roughly 15% of our client accounts who experienced degraded performance and short pockets of intermittent downtime. It has since been resolved but we'll need to do a closer look at the data during US office hours for a more complete and accurate picture.

An update on extensions

Anything significant will be for v5. We have two code bases which have diverged significantly, so it's a fair bit of work to merge v4 code into v5. I currently do this manually with patch files.

We've released some news on Invision Community Five, but third party developers won't have enough information to get started on updating their apps. We're at the peak of the mountain now and ready to start showing more in the coming weeks. I am really excited to start showing off the culmination of just under six month's work from our team.

No dates, but we're unlikely to release a major new version around a public holiday when our team are enjoying time away with their family. You'll definitely see more of Invision Community 5 this year. In terms of the JS bundling changes, the answer is 'no' but not 'never'. We want Invision Community 5 to be the start of something new, and we'll be converting JS over to native and changing how things are bundled over the early part of its lifecycle. We didn't want to add another 6-9 months onto its development schedule to rewrite all the JS which is a massive undertaking and offers very slim benefits to almost everyone outside of Google speed tests.

Hey, the 80s was peak civilisation.

For cloud, we may be able to do this by the realtime system to see who is on the page and actively using the editor.

Automated using various metrics. Yep.

How’s the music on this one @Dll?

This is correct. We have a new license for v5 which alters the renewal structure and removes the ability to "mix and match" apps. To download v5, you'd need to first switch licenses via the client centre. To be clear, you won't need to re-purchase a new license, just agree to switch to the v5 license.

Thanks, we're currently investigating.

If you go to Settings > Advanced Configuration and under the tasks block, click on "View Tasks" in the description, you should see the view updates task with a lock next to it, click that and then it should run ok.

For caching to be effective, it needs to be more a long enough period of time to make it worth storing, but short enough that content is updated regularly. A 30 second cache will increase overhead but for very little benefit. We have found that the 15 minute cache is the best balance between overhead and usefulness. We have "realtime" activity for cloud customers so you can always see who is active in a topic regardless of caching and when members log in, they are served live pages, not the cached ones.

Glad it’s all resolved now. I’m going to lock this now as it’s all sorted.

Cloud customers do get tickets. The ticket was sent in around 1am our time Sunday morning. I responded around 8:30am and got the help of Marc to investigate the issue. We have auto renewing SSL certificates which rely on a CNAME being added to the domain which we inform the customer at the time they sign up. On this occasion the CNAME is missing from the domain name which we do not control so the auto renewal is flagged as PENDING on our side. As soon as the CNAME entry has been added, it’ll auto renew. So to be clear, the customer got support on a Sunday morning by two members of our team for an issue that was with the domain name we do not control.

The Hive team is also aware of the logo issue and is working with FZ to diagnose the issue.