https://issues.apache.org/bugzilla/show_bug.cgi?id=56035
--- Comment #17 from Eric Covener <[email protected]> --- (In reply to Yann Ylavic from comment #16) > (In reply to Eric Covener from comment #15) > > Current rev of patch has a problem with access_status=AP_FILTER_ERROR, > > ap_die will copy it to r->status, so the caller cannot restore it. Left at > > a negative value, it's logged as -. > > I don't see how, in the proposed patch, ap_die() can set r->status to > anything than a valid HTTP status (AP_FILTER_ERROR, AP_NOBODY_READ/WROTE or > whatever). > > When the given access_status is not a valid HTTP status (as per > ap_is_HTTP_VALID_RESPONSE, negative values included), either ap_die() does > nothing (response already sent), or it is forced to > HTTP_INTERNAL_SERVER_ERROR (and so will be r->status below). > > Do you have a path for this to happen? You're right, I had to hand-merge parts into 2.2.x and looking now it is mangled -- You are receiving this mail because: You are the assignee for the bug. --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
