anyone have any idea why my sql server keeps stopping. I have a server with just one site, not much traffic, and a sql server. What can I do to insure this doesn't happen again?
.John

Hi!
Is it mysql or M$ sql or Postgresql.
Generally in mysql you must see the following
1) The disk space
2) The pid
3) The logs to find the error.

Have a nice day
Regards
Amar

Have you checked the error log for the server?

Frank

MS SQL, disk space is ok.

Windows generally there is just one thing I generally say "May god be with you... "
You generally do need to trouble him for everything when it comes to windows server... :)
Sorry I could not be of any help here...

Regards
Amar

interactive6,

MS SQL Server still has an error log. It's ususally pretty good about telling you a few things before it crashes.

Frank

does sql have its own error log?
.j

now it wont start.

i don't see anything in event viewer
.j

anyone have any steps i can take to try to fix this?
.j

It's been quite a while since I touched MS SQL server, but..

Look in the directory where the software is installed, then poke around for it.

Frank

Look for a folder named LOG and a file called ERRORLOG.
It will show SQL server error messages.

Or just use the enterprise manager open the Management folder and the SQL Logs folder.

- Apply the latest service pack.
- Look in the error log and event log.

Does it die right after your start the service or after a while?

Look at the memory consumption in performance monitor. How much is it?

Run the profiler and when it dies, look at the last sql statments logged. They might give a clue.

At least a crashed M$ sql won't do any harm :)

Your CPU pegged on the box? See this article http://slashdot.org/article.pl?sid=03/01/25/1245206, patch your server, and close un-necessary ports.

Dan

DENY ALL TCP 1433
DENY ALL UDP 1434

Even if you patch up sql 7, i can take out your sql server if the ports open, it'll still die with the hello dos.

Is there any technical reason a provider should be down for over 12 hours due to this problem? It seems like an easy fix to me...

Symantec articles:
http://securityresponse.symantec.com/avcenter/venc/data/w32.sqlexp.worm.html

Yes, the poorly secured ASP/ISP have a big problem. All the worm'ed out nt servers are not broadcasting balls to the walls, which will cripple the switches.

Simply blocking the ports on main routers may filter the external traffic, but now you have 100 NT servers pushing 10-45mbps each onto a poor low end cisco switch.

It's an internal problem. They must stop the sql services on the servers, remove the worm, then patch the boxes up, and somehow keep the customer happy and prevent this from happening again.

Of course if you would have simply PATCHED/filtered your box oh, over a year ago, this would not have impacted anyone.

Alas it separates the kiddy hosting companies from those who take security seriously.

Think about that poor 1 or 2 sys admins who work the weekend shift who have to deal with hundreds of owned boxes. they are probably pissing their pants trying to get their stuff up..

lol.