UI is taking longer and longer to load

[Arasaka]

Greetings,

I have noticed that the Admin page has been taking longer and longer to load over the last few months, almost a full minute to load or to change views. My configuration is pretty sold with very limited over the years changes.

The attached screenshot shows my system is not at all stressed and Log Retention is set to 14 and is sitting at 423.7 MB. I am stumped as to why the UI Admin page is taking so long to load.


Screenshot 2022-01-11 083044.jpg

[dashpuppy]

Sounds like Hard drive related.

[dashpuppy]

The UI on a computer screen is one of the most important parts of the user experience. It gives us the ability to interact with our computers and with each other. As we get more connected to the internet, it becomes more important for us to have a fast and smooth UI that works on all devices. This can be solved by implementing intelligent loading algorithms that are able to detect what kind of content is being loaded and how it needs to be displayed.

Huh ...

[sky-knight]

Huh ...

Ignore the bot, it'll start spamming insanity soon and be banned.

[dashpuppy]

Ignore the bot, it'll start spamming insanity soon and be banned.

Exactly what i thought it was. Tnx sir.

[nicholasb]

I realize this is over a month old, but did you ever figure anything out? In June 2021, after being frustrated with the slowness, we setup a rule to bypass the UI traffic and found a substantial increase in performance; it was very tolerable. But around December 2021, after the 14.4 update was released (as best as I can tell), the interface slowed to a crawl again, but only if you access it from the WAN or via command center. If you access it via the LAN, it's very performant. We had some theories, and so to test them, we setup a test Untangle and setup an ad-hoc to the WAN and it was sluggish, so there seems to be something on the WAN side, that by default and out of our control, is causing the slowness, and because we were directly connected to the WAN via network cable, it can't be blamed on Internet speeds or latency.

We have 12 Untangle's and this impacts every one of them. I guess it is time to submit a support ticket.

[sky-knight]

I have one box that does this, and I've never found a way to improve it. It's only on the dashboard though, once I get off the dashboard things are back to normal.

Also, all traffic that terminates on Untangle itself is automatically in "bypass" state, so your bypass rule doesn't do anything. Which only adds a layer of further mystery to this situation.

[nicholasb]

We actually found that, at the time of our "fix", outgoing traffic from Untangle, that is, the return traffic from Untangle to our source via browser, was showing up in the IPS logs and application control logs, and that's why we put the rule in place (albeit with the appropriate IP, protocol, and port restrictions). That traffic then started to appear in the bypass logs. You are correct, however, that seems to have been a bug or something else, because I look now and that same traffic isn't logged anywhere, not even on the bypass tab.

Removing our bypass rule made no change. I'll open a ticket with support and try not to be the guy who doesn't share the solution.

[nicholasb]

I just realized that somehow our skins got defaulted to Modern Rack when our default is to change them to Classic Rack; maybe that change happened during an upgrade? I switched them back to our preferred Classic Rack, and we are, amazingly, back to normal. We tried it on 4 Untangle's so far, and just changing the skin seems to have "fixed" it for us.

[sky-knight]

My problem child didn't change moving from Simple Gray (the current default), to Classic Rack or back again.