Ryan Ashbrook

Yes, this comes standard as a part of the 4.7.1 update. 🙂 

We're aware of this issue and working on resolving it.

Go to your Admin CP, click the bell in the top right, then click Notification Settings on the top right of the menu.

On that page, look for this section, and select either "Yes - once until addressed" or "Yes - every occurrence", depending on your preference.

Hmm... I'll make a note of this. InnoDB is always preferred because it uses row level locking, rather than table level locking like MyISAM. I see forums_posts is in the MyISAM column when that's probably the most important table (outside of core_search_index) to be InnoDB.

A fix for this will be included in 4.7.2.

A fix for this will be included in 4.7.2.

In addition to that,  instead of using \IPS\Session\Front::i(); for external scripts you should use \IPS\Dispatcher\External::i();

1 hour ago, sobrenome said:

Is there an IPS4 setting to add metadata (like "Cache-Control max-age=31536000") when we use S3 as file system?

You would likely have better control over things like this if you place your S3 bucket behind a Cloudfront distribution. It's a bit complex, but you can achieve some very specific caching situations.

Just now, Nathan Explosion said:

BOOM!!!

Yes, this is fixed in 4.7.2. 🙂 

That is a legacy column that is no longer used since the latest versions of IPS4 use the password_hash function. That column is still referenced, however, if a user has not logged in since an upgrade from IPB3, so the system can update the password hash to the new version.

1 hour ago, DawPi said:

Hm, load app info every page check isn't much efficient. Maybe someone else has better idea.

But thanks for sharing it!

For what it's worth, the current application and module is already loaded by the dispatcher (as mentioned by CodingJungle), so it won't reload it - it would just use the existing data already in memory.

Additionally, applications and modules are cached in the data store so loading them on the fly should have very little effect on performance.

As Marc mentioned, this is likely an out of date theme. Specifically, the forums > topics > topic template. If you open that template in your theme, click the Variables button and make sure it looks like this:

$topic, $comments, $question=NULL, $votes=array(), $nextUnread=NULL, $pagination=NULL, $topicVotes=array()

Try now, I believe we have resolved the issue.

Your community is appearing online for me, currently. What specifically are you seeing?

We did have a brief outage earlier that has since been resolved. You can see the current status (and subscribe for future notifications) at https://status.invisioncommunity.com

Only Forums, Blogs, and custom fields in certain areas support Polls at this time.

Yes, that is legitimate code used by Facebook Pixel. You can learn more about that here. It's harmless, and doesn't actually show on the site itself unless you specifically enable it.

Since you appear to be using template disk caching (which is why you found that in a file in the uploads directory), if you delete the file then it will simply automatically regenerate.

This is kind of a difficult scenario - technically, they are in fact two different pages (and you can see this when you look at the View a category button on the right, above the entry list). The first one is the overall blog, which would show every single entry. The second is viewing a specific category within that blog, which would only show entries in that specific category.

If that user posted another entry within that blog, but in a different category, the issue would resolve itself because the first link would show both entries, whereas the second would continue to show just the one.

So then you have the opposite problem, in this scenario, with the canonical link pointing to the root blog while viewing a category, instead of pointing to the category itself.

Just now, sibomots said:

For all members?

If you click the link to Recount, it will give you the option to do either that individual member, or all members.

At the end of the day, I'm not sure it really matters, so whichever you feel is best will work fine.

The only table that would have been created during upgrade would have been the calendar_events table - and technically, it wasn't actually created but rather renamed from the old cal_events table for the Calendar application.

I am unsure of the other two tables - the first one looks like a third party application, and the second one looks to have been a copy of the posts table (which was renamed to forums_posts). Searching through both IPS4 and IP.Board 3, and there is no reference at all to any "posts2" table.

Regardless of where they came from, it is fine to simply change them to InnoDB.

Currently taking bets on how many times @Ehren is mentioned in this topic.

11 minutes ago, Randy Calvert said:

IPB does not have any sort of direct integration path, however there looks to be options through Zapier. 

https://zapier.com/apps/monday/integrations/invision-community

We actually do this ourselves for various things (though we do it the other way around - from here to Monday). 🙂 

You can change this via the language system. ACP > Customization > Languages. Click translate on your language and search for the following and adjust accordingly:

• num_short_thousand
• num_short_million
• num_short_billion

19 hours ago, Jordan Miller said:

Ahh so maybe if the bionic reading font was smaller it might help increase the speed? 

Yes, and also more subtlety between the parts that are bold and the parts that aren't. This is the example I saw, which does actually work for me.

35 minutes ago, nodle said:

Maybe it's just me, but the bionic reading did not change my reading pattern at all. I just ended up reading it at normal speed. In fact maybe even slowly since the darker contrast to lighter contrast makes you pause more while reading. I understand what they are aiming for, but I preferred the normal paragraph in the link above. Maybe it's just me though. 

1 minute ago, Markus Jung said:

It took me longer to read and understand what I just read. 

This is interesting, because I'm actually experiencing the same in this instance, but when I came across this the other day I actually did read the example text faster. I wonder if weight and size somehow contribute?