bastecutfold

I figured. My internet service is fine—I’m currently using three devices simultaneously with zero speed or connection issues. Has to be hosting. I have Bluehost, and their customer service is absolutely useless, so contacting them gets me nowhere. Probably time to finally figure out how to switch to someone better.

@macmanx Thank you! I was able to delete the Monsterinsights Google Analytics plugin, which is the only one I’ve updated in quite awhile, so I thought that might be the issue. As of yesterday I’m able to get into the backend, but it’s still slower than usual and I’m getting the “Connection Lost” error 3-4 times/hour. Does that mean it’s most likely a hosting issue? I don’t think it’s my internet connection as speed on all other websites seems fine, and even when the WordPress connection is lost I’m still connected everywhere else.

Thank you, but as I mentioned this isn’t a general optimization issue—it’s a drastic change that happened overnight, so I’m trying to figure out why. I’ve already done quite a lot of work on optimizing my images, and until this changed happened a few days ago, speed was good and the site and backend loaded quickly and without issue.

Also at the moment I’m also unable to load the backend in order to do any additional optimization.

Sure. How will I know if it’s causing the error? Will I receive some kind of warning, or do I need to schedule a test post after each one?

I’m just using the native WordPress scheduling, no plugins.

For settings, the following are enabled:

Page cache
Validate timestamps
Object cache
Browser cache

@vmarko I tried this just now, and with W3 Total Cache disabled, the test post published as scheduled. So it does seem the issue is related to the plugin.

Thank you. Yes, I’ve always had the warning in site health about “a scheduled event has failed,” but the scheduled posts have always worked anyway.

I’m now seeing a new warning that says “outdated SQL server,” which seems to have appeared within the last few days. This is also when scheduled posts stopped working, so I assume they’re connected. It says I need to contact my host so I’ll do that.

In the meantime, is there any way to manually publish my scheduled posts from the past week, short of copy/pasting all of them into a fresh post?

Ah, so the image you’ve linked is the “featured image” for my theme. Every time I post a new post, this image automatically changes to display an image from my most recent post.

Sounds like that means the lazy loading issue is within the theme? I’ll try reaching out to the developer to see if they can help.

Thank you! All of this is a little bit over my head, so if you don’t mind a few follow up questions…

Largest Contentful Paint image was lazily loaded – I can see that the Lazyloading is not enabled in the W3 Total cache, so this means that the Lazyload is enabled somewhere, and this may be some other plugin or a theme.

How would I check which plugin uses lazy load? In trying to improve speed I’ve deleted all but the essential plugins so I only have a few. I’ve already deleted Jetpack and Smush, which were the two I thought were causing the conflict, but it must be something else?

Largest Contentful Paint element – This image (the same one for lazyloading) is the LCP element and therefore should not be lazuloaded

Is this one specific image on my entire website? Or one per post? How would I determine which image is the largest contentful paint element?


Serve images in next-gen formats – This means converting the images to webp format. This is also available in the W3 Total Cache and you can enable this option in the General settings

I already have this box checked in settings, do I need to do something else to enable it?

Thank you! Yes, I’ve deleted Smush. I also removed Jetpack and all other non-essential plugins, so I’m down to the bare minimum. I also used Site Health in troubleshooting mode and as far as I can tell the issue is possibly W3 Total Cache related. Bluehost told me to use this plugin, but now that it’s installed, activating/deactivating doesn’t seem to make much difference. Images do seem to load a little bit faster if I disable lazy loading, but Site Kit is still telling me that my page load time is around 4.5 seconds, or “poor.”

I’m going to work on scaling down my images, and actually just heard back from the theme developer with the best sizes to use. But something else is definitely still at play, because even with the large images the site has never taken this long to load. @vmarko, could it possibly something related to something I need to change in W3 Total Cache settings?

Yes, I think moving to a new hosting supplier might be in the cards. Short-term I just want to get the site speed back to where it was, but longer term it’s something I need to investigate.

Between chat and phone I think I spoke to five different people yesterday, and really accomplished nothing. I’ve been with Bluehost for more than a decade, hosting multiple sites for business and personal projects, and their customer support used to be truly top-notch. Sadly that’s no longer the case. I got better advice for free from these forums and Reddit than from the company I’m paying to help me.

I’ve now been upgraded to “level two” and am supposed to receive an email from a technician today, so maybe that will be more productive!

Sigh. Still on the phone with them. They just keep saying they can’t replicate the issue and there is nothing they can do but clear the cache again.

Just got off the phone with them. They told me to hire a developer to fix it, but unfortunately that isn’t in the budget right now.

I was then placed on hold for about 30 minutes while they tried to troubleshoot. They cleared the servers, and told me to wait 1 hour and then call back if nothing changes. Currently waiting 1 hour but not overly optimistic, considering this is what was tried earlier this morning and it didn’t fix anything.

After my own Googling, I actually suspect it’s a caching issue—I found quite a few posts from people saying W3 Total Cache caused slow image loading and slowed down the backend, which I’m also experiencing.

If I enable W3 Total Cache, all of the warnings above disappear and my site health is “good,” but the images barely load. If I disable it and use the Bluehost cache instead, I see a slight improvement in image loading but get a critical issue: “Page cache is detected but the server response time is still too slow” and a recommended improvement: “You should use a persistent object cache.” And with either method, it’s still taking several minutes for a page to fully load.

Thank you. Unfortunately this didn’t seem to change anything. I kept seeing the same issues regardless of which plugin I checked.

However, I had already removed the Smush plugin (which I suspect was the one which caused the issue) so I couldn’t test it. Other than that and W3 Total Cache, the other plugins are all things I’ve been using for months/years without any problems.

I’m wondering if I changed something when I set up the Smush settings and that change “stuck” even though I’ve disabled the plugin? In addition to speed, the images load differently now (they sort of “click” or “unfold” one section at a time) so I’m thinking I changed this somewhere but have no idea how to change it back.

And yes, Smush was just installed yesterday. I kept seeing the recommendation to optimize images as a way to improve speed, and this plugin had great reviews so I thought it would do the trick.

However, before yesterday my site was on the slower side, but fully functional, and all images loaded just fine. Now half of them don’t load at all, or they load strangely (wrong size, only half of an image, etc.) despite looking fine in the backend.