mychiefs58

tom,

.container-12 keeps the mega panel within the bounds of the overall container now, although it is still restricted to the width of the menu itself and not the 960 px of the panel. still, it does fix the overrun on the right side and looks much better, so thanks for that.

i’ve tried two different widgets in the menu, text and my recent posts. the my recent posts is what broke it as described above. the text (which was just containing <img> tags, didn’t work at all. nothing showed up.

forgot to link my site.

thanks, Tom.

the temp fix works so i can keep playing around with it. gl on finding the issue!

nate

you bet. done.

yeah…i removed the Envira Gallery from everywhere. i didn’t have time for my website to be broken. i do like Envira, but won’t be using in on this site any longer.

James,

How is this a 5 minute troubleshooting process? I cannot duplicate the problem on demand. It happens sporadically…very sporadically. I have no idea what conditions to set in order to make it reoccur. And as our site is visited globally by engineers who are looking for million dollar laser manufacturing systems, I can’ just decide to pull it down overnight when India, China, and a dozen other manufacturing rich countries are looking at us.

If I knew what to do to reproduce it, then I’d be able to run through the plugins fairly quickly. Alas, I don’t know how.

As for the max_size thing, I’ll ask WebSynthesis what the setting is, but the post I was saving yesterday was maybe 500 words. I doubt that would break that limit.

If you have an idea how to make this a 5 minute exercise, I’m all ears.

Sorry for the snarky tone yesterday. I was pretty aggravated.

This happens to me about once a month. My plugins are kept to a minimum on this site and the theme has been pretty solid for us since installing it a year ago.

It is impossible for me to go through the processes above. This is a live work site and the owner will not pay for or provide resources for a dev site. That’s another level of aggravation altogether.

I get believing this to not be a core issue. But, to be fair, that’s the same thing everyone gets when they call any customer service line. “It’s not us…it’s _____.” In the end, whether or not it is doesn’t really matter to me, or to the thousands of other people who have experienced this. We’d just like it not to happen.

With that many plugins (and I’m building one myself) and so many million sites, it would seem likely that it’s not just one or ten plugins causing this, unless it were one of the very few used by several million people across genres (SEO/Caching/etc). That leads me to believe that there might be many, many plugins causing this issue, which might mean authors need a better understanding of how something works. Obviously the same is true for themes.

I cannot speak to other’s experiences with this error, but as I get it occasionally, while having a pretty limited list of things I do here…write/add media/that’s about it…I’d be hard pressed to duplicate the issue on purpose. It just happens once in a while. Always while updating a page or post, that much is true, but not with any sort of pattern (I’m a guy who notices patterns). In this particular instance I was in WP all day long with no issues and then…

Perhaps I just don’t understand the process well enough (I’m sure I don’t) but isn’t there some way to hold that post data during the submission process and give an error in a way that keeps it? Maybe like a form that doesn’t submit so you hit Back and the fields are all still populated? It just seems that there can be a more elegant method of the process that might help mitigate this for everyone.

whilst trying to troubleshoot on my own, i discovered something odd.

if i right-click on the “gallery free” 2nd tab and select Inspect with Firebug, i can sometimes get the galleries to show up…at the expense of the galleries on the first tab. seems they’ll only show in one or the other.

over the weekend i was able to get w3 cache to work on the site…or part of it. i have object, browser, and database cache enabled. next weekend i’ll look at enabling page cache, as well.

i left minify and page disabled for now. i already was running a minify plugin before i added w3 so i decided to leave it at that.

i’d hoped to test all the way through w3, but we had a 404 error anomaly that was affecting only one part of our website. i had to put most of my time into that.

still, some caching > no caching.

not sure why that would be. it’s just a direct link to the file on my server.

it looks pretty much exactly like the photo at the top of this forum page.

https://wordpress.stackexchange.com/questions/59309/page-output-in-strange-characters

if i look at the page source code, it looks the same…no change. i am reading more about improper encoding being sent being a reason for random unicode characters being inserted into otherwise normal text. i don’t know if something is changing the encoding completely through all the mess that’s leaving me with that page.

i installed wordfence on my personal website, hosted in the UK, and have had no issues with running the Falcon Engine there. here is a link to that server configuration. perhaps someone who knows more than i can compare it to the one above from our Bluehost server and see if anything can be identified as a cause there.

https://www.cmslaser.com/wp-content/uploads/2014/09/generated-mysite.pdf

i’m going to have to test w3 cache this weekend. there are too many settings to try one at a time. i’ll just disable them all and then bring them back one at a time until it crashes again.

i just tried disabling gzip and that didn’t seem to do anything. i’m wondering now if it could be minifying js/css twice. but…i’ll work on it this weekend.

let me put the sum of what i’m running, or not, at any given time, and what causes the problem all in one place.

my site runs about 25 plugins at this time. prior to this wingding unicode issue, the only problem i’d ever had is the All In One SEO program somehow causing occasional 404 errors on pages that were valid. It took some digging through my error log to start narrowing that down, but once the trail was sniffed, it did lead me to that plugin, and disabling (and deleting) it solved that.

wordfence had been installed as a security plugin, along with ithemes security for about 6 weeks. while digging on the AIO SEO program, i decided to dump ithemes security, too, and just go with wordfence. so…wordfence was the only security plugin running on the site for about 2 weeks. it does security in conjunction with our CDN company (incapsula) and i’ve been pleased with the results.

while looking through wordfence one day, i ran across the Performance Setup again, which i’d seen when installing but had left alone while getting all the bugs out of the site. i decided to turn on the Falcon Engine at that time. no other caching plugins were present at that time. the site went down, unbeknownst to me as i did that just before leaving work.

the next morning, the solving and return of the problem led me to the Falcon Engine and we left it off. as wordfence states that the Falcon Engine will not work in conjunction with W3 Cache, i decided to try that plugin with the performance part of wordfence turned off. it brought the same result.

i disabled W3 and we left that alone for the remainder of the day yesterday. this morning, from home, i set wordfence to basic caching to see if that worked. i saw nothing improper in the half hour before i left for work. after an hour drive, and approximately a half hour into my work day (2 hours from the change) the problem was back, fixed by removing the basic cache setting.

the StackExchange link posted by Tim above seems to have plenty of evidence (or as much as i can find anywhere about this problem) pointing to cache plugins and specifically to gzip issues. i am not anywhere near versed enough in the complexities of how gzip works, how the headers are created or passed to browsers, or anything like that. it just seems to me that this might actually be the problem.

i can’t test disabling the gzip until later this evening…at least sometime after 6 pm pst.

here’s my system’s configuration as generated by wordfence, if that helps.

https://www.cmslaser.com/wp-content/uploads/2014/09/generated.pdf