On 20 April 2010 07:21, Alec Flett <[email protected]> wrote: > so it turns out having MaxRequestsPerChild set to 1000 was what was causing > our mod_wsgi daemons to die. > > we've been running a week now with MaxRequestsPerChild to 0 and haven't lost > a single daemon. > > .. so I think we're actually fine for our deployment now because we don't > need to recycle our apache children anyway (that was leftover configuration > from mod_python) but this is perhaps a bug in mod_wsgi...
The MaxRequestsPerChild directive is an Apache thing and not mod_wsgi, there shouldn't be any interaction. If there is an issue with extreme churn of Apache server child processes it would have to be within Apache code that problem is occurring. Thus, don't see how mod_wsgi could have a problem given that it is relying on Apache Runtime Library code to manage processes. All the same I will keep it in mind and if I can think of any tests to run I will do so. Thanks for letting me know of outcome. Graham > Anyway, Graham I've taken your other advice and fixed our spare server > limits/etc to reduce churn - all that stuff works well now too. We're in a > much more stable state... thanks! > > Alec > > > -- > You received this message because you are subscribed to the Google Groups > "modwsgi" group. > To post to this group, send email to [email protected]. > To unsubscribe from this group, send email to > [email protected]. > For more options, visit this group at > http://groups.google.com/group/modwsgi?hl=en. > > -- You received this message because you are subscribed to the Google Groups "modwsgi" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/modwsgi?hl=en.
