Apache's SSL broke after system hangup

W

wa2

Verified User
Joined
Apr 12, 2007
Messages
8
Hi,

were having a huge problem on a main production server after some hardware trouble. Everything seems to work fine, including Apache, except when we try to start it with SSL (httpd -k start -DSSL or service httpd restart). Apache works fine without -DSSL but we really need HTTPs for some sites (which are not working right now). Bash said:

/bin/bash: line 1: 3257 Segmentation fault /usr/sbin/httpd -k start -DSSL

A few times, with varying numbers (instead of 3257). A full fsck was already run at the filesystem. Please help, since I don't have a clue where or how to fix this.
 
What kind of system hangup did you have?

Have you tried rebuilding DirectAdmin's hosting stack through customapache or through custombuild? fsck doesn't guarantee all files are good, only that the file system is consistent.

Jeff
 
The hangup reason is unknown, the system reset itself around 5 AM leaving no trace, altough there hasn't been a power failure. I'm suspecting a full drive (the server is running in VMware and full disks aren't handled that well.

To get things up and running without having to rush to the DC, I started the vm at the failover DC. The synchronisation process was interrupted by the outage which caused some file corruption.

Eventually (don't ask why I didn't try this before) I ran an strace on apache, which told me the the file "/var/log/httpd/ssl_scache.pag" was causing the segmentation fault. Nano was able to read the file but when I cleared it, Apache started fine. Problem solved!
 
You must log in or register to reply here.
Back
Top