What causes internal server errors?

Mister Bushice · Post by **Mister Bushice** » Sat May 19, 2007 5:55 pm

I have a remote hosted mysql database. When submitting updates to it thru a php based interface, I've noticed that just recently there have been quite a lot of internal server errors popping up.

My provider never gives anything more than very generic answers ("They're working on it, it's a server problem"), or else if I submit a "trouble ticket", I get a "Problem resolved" email 48 hours later, with no explanation detail other than "server issue. problem solved" POS message.

Anyone of you guys know anything about internal server error causes? It isn't losing data, except that whatever I submit for that page doesn't go in, but subsequent submissions will work ok.

ElTaco · Post by **ElTaco** » Tue May 22, 2007 4:13 pm

I haven't even the slightest clue. You would probably need more information before you can even begin to trouble shoot it. You might call them up and instead of asking them to fix it, see if you can trouble shoot it. In other words, see if you can figure out what causes it and then they may be abel to better track it and give you the full error message. Once you have that and they have it, they may be able to fix the real problem.

Here are some links but they may be useless, without knowing the full error and what causes the error:

http://support.microsoft.com/kb/830466
http://www.c-sharpcorner.com/UploadFile ... 00ISE.aspx (probably not it)

There are more errors but those are the obviuos ones. To even begin to help, and its probably nto likely we can recommend too much, but to even start, we'd need the OS, SQL db version, and the actual Error that is thrown, including the error number and probably the text. My guess from your description is that you can't see it, but it will give an error so the other possibility is to figure out what throws it exactly, what is the code you are doing exactly when the error appears and then I would check to see if maybe you can try a different syntax, if not then call them up and have them find the actual error in their logs and throw a couple errors so they can paste it to you. Then you may be able to fix your code or work with them to fix the configuration/update to fix the problem.

Good luck

Mister Bushice · Post by **Mister Bushice** » Wed May 23, 2007 11:36 pm

The answer I got was:

"You are either overloading the CPU or

You are exceeding 4 database queries at once.

There is nothing much you can do. Maybe check your code to streamline it."

Of course I have not had any internal server errors for several years until recently and the code is the same now as it was then, so obviously it isn't the code.

PSUFAN · Post by **PSUFAN** » Thu May 24, 2007 5:06 am

sounds like you've got a shitty host.

do you know how many queries a board like this uses? Probably 25-50 per page. 4 at a time is NOT very many.

Internal server errors are on the server side. You often see them when a script times out, or if a service has been stopped, etc.

Mister Bushice · Post by **Mister Bushice** » Fri May 25, 2007 6:27 am

Yeah, PSU. I signed up for this provider based on M7s recommendation. He used them for the old TNW board and gave them high praise. they were very good for several years, but this past year they have begun to suck.

Tonight the thing is screamingly fast, but of course no one is on line to interfere with my queries.

Changing providers means a huge undertaking in code page transference. I just don't have the time or tech support to take that on right now.

Sucks, but for now I have to tolerate it.

Mister Bushice · Post by **Mister Bushice** » Sat May 26, 2007 8:10 pm

and the expected response:

I would like to inform you that our administrators checked
your account and they couldn't receive error. Could you
please provide us with url to the problematic script in
order to test further.

Now I TOLD Them I had a php driven query page, and that it performs a dozen different queries of the database depending on what I want. I TOLD them two specific to-the-minute clock times when I received internal server error messages, one of which occurred when I was actually on line with tech support. I TOLD them not a thing has been done to edit or change any of the query script in over a year, but the increase in ISEs began a few weeks ago.

logically this would mean to me they changed something, but shouldn't they at LEAST have a server log that shows when it throws an ISE?

I guess the next step is to document all of the queries that are resulting in ISEs.