nlv11281 at natlab.research.philips.com
Tue Jun 27 02:02:14 EDT 2006
I'm using a 64-bit AMD linux machine here - and I'm doing lots of MD5 checksumming without problems while processing terabytes of data. I can probably use some 64 bit intel as well, but in the hundreds of mails I got during my vacation, I lost track of the original post for this problem. Let me know how I can help. Mike Looijmans Philips Natlab / Topic Automation Deron Meranda wrote: > On 6/26/06, Nikolaus Schlemm <nikl at nikl.net> wrote: > >> [ .... cut (cat /proc/12345/map) output here--included below .... ] > > > So we're dealing with a 64-bit Linux process! > > That platform probably doesn't get nearly as much testing as 32-bit. > Are both of your broken machines 64-bit? Unfortunately I don't have > access to a 64-bit Intel machine; does anybody else on this list? > > Otherwise nothing looks obviously wrong with your process map > (no mixed-up python versions, etc.) > > You do however have a lot of PHP libraries loaded, and even > X11 is linked in there for some reason. I can't see how those could > affect the python md5 module; but is it possible for you to run a > much leaner Apache with most of the optional modules disabled > (perhaps even on a different port if needed), just to try to eliminate > some sort of library conflict. > >> > Also, please review the Apache error logs, just to make sure >> > nothing unusual is showing up in them. >> nothing unfortunately.. > > >> > And just to be clear, is all the software (python, apache, mod_python) >> > as-is from the linux distribution; or have you attempted to compile >> > or install any of the components yourself from source? >> nope - all coming straight from the distros > > > Hmmm. Just for some more information, could you get > * Which Apache MPM are you using? (If it's a threaded one, > could you change to say "prefork" and retest?) > * kernel version (cat /proc/version) > * exec shield config > cat /proc/sys/randomize_va_space > cat /proc/sys/exec-shield > * SELinux status > sestatus > > Also, I suspect that perhaps it might be time to get this report > into the bug database if you wouldn't mind. > http://issues.apache.org/jira/browse/MODPYTHON > If you need help with Jira or prefer not to use it, just let us know. > > Thanks for hanging in there Nikolaus, I know this must be frustrating.