lost_usr

Ugh…
Have been on the site for little while now.. making changes here and there…
Mostly to the layout… and then I go to edit the content of a page, and guess what..
My Dear “Connection Lost” message comes back. …

And I’m just like… WTF.

I’m still just totally lost as to where to begin looking for this problem.

Yep..
It’s back.

Have no idea what’s different..

I have not been active on the site for a few days.
Now I go to make a truly minor edit to a post (adding a link) and WP gives me the “error saving… disabled… etc. etc.” warning…

…. And the site is running slower and slower….. again…..

quite frankly…. this feels like the beginning of the end of this WP install…. I can’t think of any other “thing” that could possibly be wrong. Just WP itself.

Anyone with any better understanding or clairvoyance… I’m glad to entertain…

Bloody guts…
The thing appears to be working just fine again.

I’m going to mark this deal (ordeal) “resolved”…
But fully expect that I’ll resurrect it again when the issue returns…
As it has once already.

Maybe next time I’ll be better armed to pin it down and dig it out.

I hope.

Ok…
Switched back to the theme I was using (somewhat customed out)…
No problems.

Turned JETPACK back on…
NO PROBLEMS..
(my test posts are going right through without any ado..).

Site loads in normal time…

Turned on a couple other plugins…
NO PROBLEM…

Gahh… frustrating…

So I installed

ERROR MONITOR LOG plugin on my suddenly screaming install of WP…

dink about with a page to see if anything comes up…
NADA..

Go to a post page and futz with it.. nothing…
Click PUBLISH.. and instantly get this:

HTTP Request
POST (non-blocking)
https://www.noahhuber.com/wp-cron.php
?doing_wp_cron=1407658463.1918120384216308593750

Response
Operation timed out after 1005 milliseconds with 0 bytes received

Transport
curl

Call Stack
WP_Http->request()
WP_Http->post()
wp_remote_post()
spawn_cron()
wp_cron()
do_action(‘init’)

Component
Core

Timeout
0.01

Time
1.0110

Do I have this right?

If I go to my host folders… grab my media folder,
Move it up and out of my root.

Go to the DB and destroy it (erase it)…

Reinstall a new DB instance on the host, and link it to my domain… then install WP to that root folder…

Then overwrite the newly installed media folder with my old media folder…

AND then go in to the WP installation and upload the .xml file I just downloaded from my current (broken) install…

Would THAT lead to all of my past posts still “finding” the associated media files? OR would each and every one of my 2590 images suddenly not associate with the posts from the past 7 years that call on them…?

Getting into the MEDIA library – smooth as silk.

Whatever happened, it got purged after … nearly half an hour.

Get this – now (just now) I tap on the open tab that’s been pointing to the dashboard on my site for the last (uhhh.. half hour?) and it’s loaded. And running. And running pretty smoothly!!

So – I’ve just begun to export (download) my pages and posts…

and the XML is complete. (I KNOW I’m not getting all of the media, however)…. which I still need to grab.

Ok… another interesting tid-bit.
A subdomain on the website that is in question is running smoothly…
It is also running 3.9.2 … and has a bunch of plugins… and no issue.

I have JETPACK installed on the subdomain website.. and it’s connected and running…
So all things being equal, it is not JETPACK that seems to be the issue.

3.9.2 running on both…
Virtually the same plugins running on both…
One site runs fine. The other is off the air.

…. I’m officially stumped.
Something is clearly broken about WP install on one site, while another (5!) are clearly not having any issues whatsoever.

I don’t do much (read – none) editing of the source code for the WP install… neither the PHP nor of the actual DB.

Still, I guess something changed about the parameters of my install which is not overwritten when I reinstall 3.9.2.

SO – it looks like it’s something with the calls on the DB itself??

I can either wait around to find out, or I can try to back up the contents of the DB without the structures of the DB (is there a way to selectively grab and keep paired certain fields while dumping others?) ….

I’d want the post meta, the post content, the media, and media meta… but little else.
At this point, I think rebuilding the WP install from scratch sounds like the easier fix.

I’ve installed and activated Automatic’s Twenty Fourteen theme… while doing all the monkeying around on the posts page.

Clearly the idea that there is no connection is pure rubbish.

I have to say, though. It seems like the more stuff I do in the admin side of things, the slower the website gets as a whole.

Just tried sending a ping to my domain.
The response time was 112 ms for all requests.

Meanwhile – my site will no longer load on this browser, or any other browser I try.

F* Great…

So – to make this into a teachable moment – whatever is happening, it gets compounded. It gets compounded, and although the host is up, the WP install cannot resolve the issue at all after some quantity of activity.

I guess the question would be – where would I look to see the internal error logs for my install?

I can dig into the DB via my host’s FTP… although I don’t know my way around for S*IT.

Next question – wonder how long I’ve got to leave the WP install alone before whatever problem it is currently having gets flushed from the process and everything starts running normally again (until I get back in there and start trying to write posts, and WP starts trying to autosave those developing posts)…..

Access to the Media files is also painfully slow…

I’ve successfully opened another “new post” window.
Waited a few minutes to do anything at all… the window idled beautifully without giving me any error message.

Added a title for the post.
NO error message.

Added a TAG for the post
NO error message.

Added a Category for the post
NO error message.

Added a Featured Image to the post
ERROR MESSAGE received (the same “Connection lost. Saving has been disabled until you’re reconnected. We’re backing up this post in your browser, just in case.”)

Hadn’t even touched the body of the post yet.

Again, it seem pretty clear that what triggers the error to become visible is the need to save the ongoing chances to the post… i.e. the autosave function seems to be the culprit. Seems … to me… someone who’s almost illiterate and who couldn’t tell you which way on the HTML page the “header” goes.

A few more details:

The problem is not the host.
I have another 5 websites (all wp, all 3.9.2) on this same host, and they are all operating exactly as they normally do.

I switched to a different browser. Typed in my URL.
Tooks waaaaay longer than normal to pull it up.

Working my way back in to the admin panel… took almost two minutes to log in.
When I got to the posts panel I see that I have TWO drafts of my test post (testing to see if I could get the issue to go away, or how long it took it to show up at the top of my new post page once I’d begun a new post)….

My next step is going to be switching to 2014 by Automatic (the theme)…
Just to continue to eliminate every single “add on” or “option” there is to eliminate.

Again – still no plugins running.

OK… so while I’m waiting on a second tab to load the THEMES page, I opened one of my test posts, draft. It brought up my full page of Lorem Ipsum… and I’m watching it… watching it… nothing’s happening… nothing’s happening…. then I click into the body of the post and instantly the Error message pops up telling me I’ve lost the connection and that I can’t save… yada yada…

Makes sense, yes. No need to save any changes until a change is made.
BUT – it is also very clear that the issue seems to be that the NEW or Newly edited post cannot be saved to the AUTOSAVE process.

After all, no issue shows up until you DO need to save through the Autosave …
I’m going to try this again.

IN THE MEANTIME – I should point out, not withstanding that my entire WP deployment is slow as molasses … if my host or DB were TRULY not reachable, then I would be unable to make ANY changes to my site. Meanwhile, I’ve turned off all the plugins, and am now changing the theme.. and I’ve reinstalled WP twice.. just for good measure.

So, clearly the “Connection” has not been lost. There is just some BS script running somewhere that is not terminating appropriately, and is gumming up the works like heck.

Hm…..
Yep.

I have the exact same issue.

I’ve updated to the latest WP (my admin panel is now so slooooowww I can’t wait around for another few minutes to see what the 3.9# is on this date).

Things got better after I did nothing on my website for a week (not an option for most), and after I did not post for that same amount of time.

I had disabled Jetpack (it is currently the ONLY plugin operating), and while that made no immediate difference to the site ( after re-enabling it I went in and did some editing on older posts), once I began trying to post a NEW post, the error message popped back up again.

Guess I’ll rinse and repeat…
Disable Jetpack and see how things are running tomorrow.

10-4.
Thank you.

RIP Hemingway

Yes, quite the ancient thing.

Here is the support forum.

https://support.warpspire.com/discussions/hemingway

Judging from your replies – it’s either fix it myself, or find one that suits and move on?
Does that sound about right?