Hello,

Well, this is a kinda weird problem which bugs me for the last 2 days. My current server provider (www.prioritycolocation.com) has some network problems lately, and had some unexpected downtime. 2 days ago, I tought there were network problems - the server didn't respond to ping packages for a few minutes. After that, it did respond to ping packages. I was happy again - until I found out no service responded - web server didn't respond, ftp server didn't respond, ssh server didn't respond, etc etc. So basically, it seemed like the server did reply on ICMP packages, but wasn't passing thru any TCP packages.

Anyway, I rebooted my server using the APC, and it was online again. I tought it was a one-time thing, but after 1.5 hour the same thing happened again! Only responding to ping, but nothing else. I rebooted it again, and talked to a techy, and his first tought was that it was a software problem, that my eth0 was bound to the wrong device. When I clearly explained that I barely did anything to the server except hosting 1 website with PHP scripts, he tought that was unlikely too. After that, he didn't respond anymore.

The server went fine for another 24 hours or sth, but when I woke up this morning, I found my server offline for 15 hours. I had to reboot it again, and now it's up again.

However, this makes me worry VERY much! 15 hours of downtime isn't nothing, and I don't think my host's uptime guarantee covers this. However, I need some advice from people how likely this is to be a software problem or a hardware failure. What I do know is that I did NOTHING to the server configuration, I only installed MySQL, Apache, and ProFTPD, so it seems very unlikely to me that *I* did something wrong.

What do you "pro"'s think about this? Any idea what might be causing this, or what I can do about this? What I do know is that if this problem goes on, I'm getting rid of this server, since this is no way for a server to behave.

Thanks in advance for any replies.

Regards,

Leon Mergen

P.S.: The last thing that happened according the system logs before it went down was me logging in into the FTP server.

You need to look in your log files for any errors, there is bound to be useful info in them. make sure chkservd is running:
/etc/rc.d/init.d/chkservd

Is this a cPanel/WHM server???

It's a plain FreeBSD 4.5 box, with installed from ports :

- Apache
- MySQL
- PHP
- ProFTPD

And that's all that's installed. So that's basically the reason I don't think it's a software issue - there is no software that could cause it.

I would say bad memory, but it resonds to pings, why dont you ask to get a install of the OS?

Off topic but, a good idea...

Wouldn't it be nice to have a UPS that includes interfaces w/ your network. You assign it the IP or name of the machine that it is plugged into. Every X minutes/seconds it will ping or wget or ftp, which ever protocal the machine serves... If it gets a failed response X number of times it reboots?...

If this exists can ya point me too it (via PM to keep thread on topic)?