Received: by taz.hyperreal.com (8.6.12/8.6.5) id EAA19878; Tue, 11 Jun 1996 04:50:55 -0700 Received: from life.ai.mit.edu by taz.hyperreal.com (8.6.12/8.6.5) with SMTP id EAA19865; Tue, 11 Jun 1996 04:50:52 -0700 Received: from volterra.ai.mit.edu by life.ai.mit.edu (4.1/AI-4.10) for new-httpd@hyperreal.com id AA00599; Tue, 11 Jun 96 07:50:50 EDT From: rst@ai.mit.edu (Robert S. Thau) Received: by volterra.ai.mit.edu (8.6.12/AI-4.10) id HAA18610; Tue, 11 Jun 1996 07:50:48 -0400 Date: Tue, 11 Jun 1996 07:50:48 -0400 Message-Id: <199606111150.HAA18610@volterra.ai.mit.edu> To: new-httpd@hyperreal.com Subject: Re: WWW Form Bug Report: "srm.conf-dist has invalid Meta* lines" on Linux (fwd) Sender: owner-new-httpd@apache.org Precedence: bulk Reply-To: new-httpd@hyperreal.com Hmm... any way we can specify something like "unknown directives within are considered fatal, otherwise considered nonfatal but complained about loudly"? If it was obvious to me that misspelled directives in were the only ones that were truly dangerous, I'd be happy with this. However, other cases are also worrisome --- "Optiions" and "AllowOveride" in particular. I really do think it's preferable to selectively allow the server to continue in cases where it can be expected to fail safe than to try to enumerate all the dangerous cases.... (The sort of thing I'd be happy with is having syntax errors in a section disable access to that directory, and syntax errors in a disable that particular virtual server, without killing off the whole thing). rst