[imp] Re: Anyone try eaccelerator?

liam hoekenga liamr at umich.edu
Wed Apr 20 13:27:36 PDT 2005


> FYI, phpa also works well with horde.  I've ran it on a production 
> system for a few years. http://www.php-accelerator.co.uk/

We've run phpa as well and it worked well, but it seems to be abandonware.  I
tried to contact the author recently, to no avail.  So... we switched to turck
mmcache, and that's been pretty nice, but we get occasionally get apache
children that get stuck segfaulting.  If you strace them, that's all they do..
segfault.  Well, that and drive up load.  It goes a little something 
like this:

--- SIGSEGV (Segmentation fault) @ 0 (0) ---
sigreturn()                             = ? (mask now [RTMIN])
--- SIGSEGV (Segmentation fault) @ 0 (0) ---
sigreturn()                             = ? (mask now [RTMIN])
--- SIGSEGV (Segmentation fault) @ 0 (0) ---
sigreturn()                             = ? (mask now [RTMIN])
--- SIGSEGV (Segmentation fault) @ 0 (0) ---
sigreturn()                             = ? (mask now [RTMIN])
--- SIGSEGV (Segmentation fault) @ 0 (0) ---
sigreturn()                             = ? (mask now [RTMIN])


We do clear the cache everytime apache /etc/init.d/apache is accessed.

Has anyone else had this problem?

Does eaccelerator do this?

Liam


More information about the imp mailing list