APLawrence.com -  Resources for Unix and Linux Systems, Bloggers and the self-employed

Kerio Connect crash dumps and the kassist tool


2012/09/15

I've never had my Kerio server crash, but it has happened to at least one or two customers who I support. Kerio is, of course, keenly interested in preventing such things from ever happening and obviously such bad behavior is not something you want either. Fortunately, these types of problems are rare.

If your server should crash, it will almost always restart itself automatically. On Windows systems, the machine that Kerio is running on will inform you of the crash. On Linux and Mac OS X, you might not even be aware that there was a crash if you had no reason to log in to administration - that's where you'll learn that there was a problem.

What happens in both cases is that you are offered an opportunity to send the crash dump (a record of what happened just before the crash) to Kerio. It's actually important for you to do so, but there are some things you need to know.

Notice%20of%20crash%20dump%20on%20Linux%20server

By the way, I deliberately caused that crash shown above by sending a "kill -11" to the mailserver process. I do NOT recommend you do that out of idle curiosity!

Support is not notified!

The most important thing to understand is that Kerio Support is NOT automatically notified of your crash, even if you do send Kerio the crash dump.

That may seem surprising. What's the point of sending it if Kerio Support won't know about it?

The answer is that Kerio Quality Assurance and Development are the folks who are interested in the crash. If it wasn't caused by faulty hardware or operating system goof-up, QA and Development want to dig in and see what they can do to prevent the same thing from happening to someone else.

However, if you do have an open ticket, you can put that ticket number in the description and (according to what I've been told by Kerio support) "a note will appear with a link making it easy for the support engineer to access the crash".

If you are one of my customers, you should have me open the ticket. I'll give you the ticket number to reference.

Should you automatically open a support ticket if you have a crash? No, probably not: the crash may have nothing to do with Kerio. You might have bad RAM, for example or some other hardware problem.

However, if you have frequent crashing, you should definitely open a support ticket and you should absolutely send every single incident to Kerio. As that same support engineer explained to me "squeaky wheel counts a lot on the crash dump server here".

So what's frequent? Well, Kerio might disagree, but I'd say that if it happens more than once you ought to open a ticket. Maybe it is some problem on your machine, but if it is, Kerio Support might help you know that.

Zero Byte Dumps

Crash dumps get stored in the same directory that has your configuration file. The name will vary: a Linux system may just be named "core" or it may be "core.7922" or some other number (see Controlling core files on Linux for more information). On Windows, it will be "user.dmp" and on Mac OS X, you'll find it in "/cores/core.xyz" where xyz is the PID of the process that crashed.

If there is more than one crash file, you should be able to tell by the files date stamp which one is of interest. On Linux, you can also use the "file" command:

# file core
core: ELF 32-bit LSB core file Intel 80386, version 1 (SYSV), SVR4-style, from 'rio/mailserver'
 

What if the file is 0 bytes? That can happen, and I was surprised to learn that Kerio still wants to know. That support engineer said "If they are all 0 byte uploads, that tells us something about the system and the crashes too, so just keep doing it."

Kassist

There is a command line tool that can send crash dumps also. See Using Kerio Assist tool for detailed instructions.

Keep this in mind

As I said, these crashes are rare - you might never see one. If you ever do, please refer back to this article to remind yourself of what to do, and if you feel you need to open a ticket, do that BEFORE sending the crash.

After you are sure Kerio has received it, you can delete the core file from your system to reclaim space. These core files can be quite large: the one I forced with a kill -11 was 86,540,288 bytes and they can be much larger.



Got something to add? Send me email.





(OLDER)    <- More Stuff -> (NEWER)    (NEWEST)   

Printer Friendly Version

-> -> Kerio Connect crash dumps and the kassist tool:




Increase ad revenue 50-250% with Ezoic


More Articles by

Find me on Google+

© Anthony Lawrence



Kerio Connect Mailserver

Kerio Samepage

Kerio Control Firewall

Have you tried Searching this site?

Unix/Linux/Mac OS X support by phone, email or on-site: Support Rates

This is a Unix/Linux resource website. It contains technical articles about Unix, Linux and general computing related subjects, opinion, news, help files, how-to's, tutorials and more.

Contact us





Never let a computer know you're in a hurry. (Anonymous)

That's the thing about people who think they hate computers. What they really hate is lousy programmers. (Larry Niven)












This post tagged: