500 server error after attempting to upgrade to 4.4

We may be able to help, but we’ll need a more detailed error message. Internal server errors are usually described in more detail in the server error log. If you have access to your server error log, generate the error again, note the date and time, then immediately check your server error log for any errors that occurred during that time period. If you don’t have access to your server error log, ask your hosting provider to look for you.

I see you have given us a ton of log entries, but we don’t know which one of those was recorded during the 500 error.

Thank you, Jeff. The server log I posted precisely covers the time during which the update script/program crashed, and I tried to access the dashboard.

Maybe you have clients who will wait 24 hours while you generate errors, and engage in a time-consuing online “chat” with for the people running your servers, but alas I don’t.

In the meantime I tried a manual update, following to the letter the official WordPress detailed instructions. To no avail. I still got system error 500s.

Eventually, I backed up thousands of photos (it’s a big site) plus the database, then destroyed everything in that area of the site and took out the database. After which I created a brand-new installation, then populated the brand-new database with the data from the old database, and re-uploaded the thousands of photos.

This worked, but I shouldn’t have had to do it. And the problem itself is not solved.

The server log I posted precisely covers the time during which the update script/program crashed, and I tried to access the dashboard.

No, the log you posted covers a nearly 5-hour window from 04:26:10 to 09:07:15 2015.

Exactly which line item is from the moment you say the 500 error?

The moment you see the 500 error in your browser, note the date and time, then just give us the log entry from that exact date and time please.

James, thank you for your continued interest.

The period covered in the log is the 5-hour period during which I was getting error messages. (The servers and I are in different time zones.) Every time I did something, I checked to see if I could access the dashboard, and got the error message.

The actual error would have taken place around the start of this time. I published the log because it is usually requested by people looking to help. I recognise that it is of little use..

The error occurred when the WordPress update script/program crashed. I had been looking at the dashboard when it disappeared. Trying to revive it, I got the first of the error messages (Internal System Error: 500)

There is no way that I could make the error occur again, since the update button is no longer on the dashboard (and I wouldn’t click on it if it were!)

As you will see from the end of my previous reply, I re-installed the whole WordPress setup from scratch, re-uploaded all the photos, and populated the newly installed database with the data from the previous database.

So my client’s problem is solved, as a a result of a lot of hard work on my part, but I still have no idea why the update program crashed.

As I said originally, we are using only one plugin (Akismet) and I created the theme myself. It is working perfectly well with the new installation of 4.4, and survived years of earlier upgrades—which suggests that it is compatible.

Ok, I believe we have a fundamental misunderstanding.

The 500 error wasn’t caused by the update. The update was run, something was conflicting with WordPress 4.4, and that conflict caused the 500.

Whenever you see the 500 in your browser, the conflict is triggered. That conflict is recorded in your log. That’s why I kept asking for the log entry from exactly when you trigger the 500. You view the site, the conflict is triggered, the 500 appears in the browser, and the real error is recorded in the server error

I’m glad its fixed now, but if it happens again in the future, we can help quicker if we know exactly what conflict is recorded in the log each time you see the 500. ??

James, thank you for that explanation.

While working on getting the installation back, I was thinking of the error messages as “historical”. I read them as saying, in effect, “it’s still not working.” But of course I was actually generating the error every time I refreshed the page, hoping that the Dashboard would appear. If I had stopped to think, I would have realised this.

It is even more obvious now how useless the cPanel “Error log” is, since none of this was recorded.

The real error messages in the server log may be of some general use, so as soon as I have the time, I will contact the people who run the servers and ask them to tell me what they can find for the AFPO Calpe site for the time in question.

Thank you again.

You’re welcome!