if (ret == HTTP_NOT_MODIFIED) {
> +r->status = ret;
> +return OK;
> +}
> +
> +return ret;
+1 as a practical fix.
For the wishlist: review whether we can do this more cleanly by making
ap_meets_conditions set r->status ins
nore the wagging of tongues.
We also have remaining bugfixes that *should* go in.
PR#39710 is simple enough to review, and another release without
fixing that would be a huge WTF??? I'm also part way through
reviewing Chris's mod_dbd rewrite. It's clearly an improvement on
what we have, b
a useful contribution if you want to.
But you're not ready to go solo. So why not put your contents
(rather than links to it) on the wiki, where some of the regulars
can help improve it for everyone?
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
l" without securing your call to it. That
makes it a great teacher!
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
l link.
I don't know if he offers commercial support, but I suspect
he does. He certainly has a lively Apache/Windows community
and offers free downloads of Windows binaries of Apache itself
and popular third-party modules.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
ead
of just CGI? How does it offer any advantage over CGI+suexec?
Not to mention its variants like fastcgi?
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
> and a set of separate hooks for storage + maybe a way to easily build
> end user display, I think it would be a very cool thing to include.
On a purely practical note, dropping it in as a subproject would
improve visibility and get it under version control.
/me puts "look at this" on his todo list.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
de an unnecessarily overhead (and
potentially large if the function gets used more in future)?
What problem does it solve?
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
-packaged distribution that uses cgi+suexec to serve
static files.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
ant to read why apache on debian/ubuntu presents
problems, at:
http://www.regdeveloper.co.uk/2006/11/04/apache_packages_support_vacuum/
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
ople have. Some people feel it's a bad
> design.
Why?
>Some people (like me) are not so sure what the idea behind it
> is.
To give the workers time to finish serving current requests rather
than aborting them.
We're open to patches if you have a better design.
--
Nick
er a beer, no matter how impossible
it may seem to mess it up.
Thanks:-)
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
ike:
-my $srcs_regex = '\.(' . join('|', @srcs_extensions) . ')$';
+my $srcs_regex = $ENV{'APXS_SRC_EXT'}
+ ? "\." . $ENV{'APXS_SRC_EXT'} . "$"
+ : '\.(' . join('|
ck has been positive. Against it is the complexity.
AFAICS we still have the potential issue in the /trunk/ code where a
connection
goes stale and prepared statements need to be recreated. I have a
proposal
for that, but it would be good to sync trunk with 2.2 first.
--
Nick Kew
sal is different again, and really belongs on [EMAIL PROTECTED],
not here, as it's entirely an APR decision.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
ther post.
On the other hand, removing the check might risk segfaulting in some
future update - perhaps something that gets configuration per-request
from a database and a rewritemap. I've recently had a similar issue
with another module, which started life using match rules defined in
httpd.conf, then g
ntial authnz reworking in /trunk/,
are you sure this patch is compatible with 2.2?
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
> same code in slightly different places.
OK, good. That tells me *if* I find time to look at it,
that time won't be wasted.
Nevertheless, a 2.2 patch (maybe at people.apache.org/~rederpj/)
would perhaps make it easier to review?
--
Nick Kew
Application Development w
ly more complicated issue. If anyone would like
> to help with this off-list, that would be great to avoid pestering
> here :)
It's your question that I find hard to decipher.
But from what you've said elsewhere, I wonder if mod_ssl or mod_deflate
might be a model for what you
oding on top
> of gzip.
OK, that's inherited from the existing version, both input and output
filters. I guess the moral is, we've never done more than pay
lip-service to multiple content-encodings.
I'm about to go out now, so an update will have to wait:-)
Thanks,
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
Doc happen at
the right time is usually to register them with
apr_pool_cleanup_register as soon as you create the doc.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
arately-quoted colon in the line
Content-Encoding = "Content-Encoding" ":" 1#content-coding
(also from having no recollection of ever having seen more
than one token in a real-life Content-Encoding).
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
a why not. The patch was an old one
that's been languishing in bugzilla.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
As for 2.x bugs, there are quite a few which are going to be
harder to deal with. Perhaps we want a new "Archived" status,
for PRs which have merit but which aren't going to get 'fixed'.
Particularly those with PatchAvailable.
--
Nick Kew
Application Development wi
On Thu, 2 Aug 2007 19:16:16 -0400
"Joshua Slive" <[EMAIL PROTECTED]> wrote:
> On 8/2/07, Nick Kew <[EMAIL PROTECTED]> wrote:
>
> > As for 2.x bugs, there are quite a few which are going to be
> > harder to deal with. Perhaps we want a new "Archi
mory when module start with apache, and unload
> when apache shut down ?
http://www.apachetutor.org/dev/pools
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
just the input filter - I didn't
plan to think through the extra complexities in the output:-)
If it'll get your vote, I'll change all three filters to remove
themselves and log a warning if called with a byterange. But that
feels to me like a separate patch.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
The old code may be a tiny tad confusing,
in that it runs apr_table_clear after the swap and thus at-first-glance
on the wrong table. But that doesn't affect your point.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
7;t
> > see how that is correct. If the filter cannot adjust content-range
> > appropriately it should either do nothing or fail, I'd say.
>
> I agree with this. See my additional mail from today.
And mine.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
> things).
Please refer to recent discussion, in particular
http://marc.info/?l=apache-httpd-dev&m=118622934222115&w=2
and take a look at mod_deflate in /trunk/ and STATUS in 2.2.
Any thoughts there?
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
nt-range is a semantic change, so I can't
> see how that is correct. If the filter cannot adjust content-range
> appropriately it should either do nothing or fail, I'd say.
Yep. That wants another patch, which I'll be happy to do.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
eflate
and other output filters by merging headers at the point
where the first data are passed into the output chain.
Looking to 2.4, any strong reason we shouldn't dispense with
r->content_encoding and let mod_mime & friends set the header?
--
Nick Kew
Application Developm
On Mon, 6 Aug 2007 14:10:10 -0700
"Roy T. Fielding" <[EMAIL PROTECTED]> wrote:
> On Aug 6, 2007, at 1:10 PM, Nick Kew wrote:
>
> > PRs 23287 and 42993, and recent discussion here, show up some
> > issues with handling Content-Encoding. Specifically regarding
) not taking forever?
[1] excluding the last chunk, which is already in /trunk/
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
, it's perfectly feasible. Just map your session management
to database operations (i.e. design what you need to do), then
code it.
[I presume you have a need for serverside session information
to be shared across all workers, or you wouldn't have asked.]
--
Nick Kew
Application Dev
e
> we adjust hdrs instead of hdrs2. This sounds wrong to me.
Good catch yet again. Dammit, I need a break.
Just fixed in r563803.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
ation.
--
Nick Kew
earlier
> than 2.0...
Maybe put out the announcements together, with "2.0 availability
delayed briefly due to issues with the bundled APR".
(Life would be so much simpler if dependencies were separated).
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
t removing the offending code.
My own view, which has no legal standing whatsoever, is that once something
has reached release-candidate, it's pretty much vanishingly unlikely that
the ASF would fail to stand by it, UNLESS a redistributor either
misrepresented it as a release or failed to take it down and publish a
notice after a problem was reported.
--
Nick Kew
ghtly bigger question. Can we do a better job
of keeping a record of changes likely to affect many users, and
perhaps ship a "2.2-compat" configuration? Our CHANGES is a
detailed record that few will have the stomach to read, and could
perhaps use an executive summary ("GOTCHAS&qu
Comment #10. Bug me if (in the absence of
any objections here) I haven't applied it in 24 hours.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
license
> headers in July of last year ;-)
Talking of which, a grep finds some more instances of 2006.
Some interesting instances are ap_release.h, and common.xsl
from whence it propagates to every page of documentation!
I've taken the liberty of updating them in 2.0 and 2.2,
as
return ap_proxyerror(r, HTTP_BAD_REQUEST, apr_pstrcat(p,
> + "URI cannot be parsed: ", url, NULL));
The second arg to ap_pstrcat is presented as a new arg to ap_proxyerror.
The old formatting was better.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
see it as being that simple.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
Index: modules/proxy/mod_proxy.c
===
--- modules/proxy/mod_proxy.c (revision 573827)
+++ modules/proxy/mod_
te code? Shouldn't this be placed in
> util.c?
Very likely. But that escalates it from a bugfix to an API change.
> How about using apr_pstrndup instead?
Indeed.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
a
byte-counting loop and modified it as above.
As for running past a \0 (which would imply a malformed input stream),
the caller expects a string, so the first NULL will terminate it
either way. Also either way the terminating condition is the
APR_ASCII_LF, or an input error.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
On Sun, 09 Sep 2007 11:25:26 +0200
Ruediger Pluem <[EMAIL PROTECTED]> wrote:
>
>
> On 09/09/2007 02:21 AM, Nick Kew wrote:
> > PR 41798 and many related ones (eg 39746, 38980 - both of which I've
> > closed today) show a history of incorrect URL-unescaping in
>
On Sun, 09 Sep 2007 22:00:25 +0200
Ruediger Pluem <[EMAIL PROTECTED]> wrote:
> [chop]
Thanks for the analysis. It's the insights I was looking for
together with some points I'd argue. But I need to give it more
think-time before proposing a revised patch.
--
Nick Kew
Appl
orker.
Does this open the way to a DoS? If a rewriterule[P] enables backends
to be derived from the request URI, then you're creating unlimited
numbers of workers, which may never be used. Where are the limits
on that?
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
ure need, and it could
be inviting people to break RFC2616 and caching.
Can you not achieve the same thing by removing the unwanted
vary entry with mod_headers' Header edit?
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
ith other prospective users of such a module, or taking
pot-luck with a student. But you're unlikely to prompt anyone
into developing it from scratch unless they themselves have an
existing need for it.
[1] http://apache.webthing.com/
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
d sound and worked for
cases identified in both the bug reports referenced.
A further improvement, round tuits permitting, would indeed be
to look deeper, and eliminate any duplication.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
illa/show_bug.cgi?id=34602#c16)
Can't speak for others, but I simply missed it.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
ction: Keep-Alive
Content-Type: application/x-tar
An application that understands tar may unpack that.
Does a Content-Disposition header help with IE7?
And would it help browsers if we supply a Content-MD5 header?
--
Nick Kew
On Sun, 9 Sep 2007 01:21:29 +0100
Nick Kew <[EMAIL PROTECTED]> wrote:
> PR 41798 and many related ones (eg 39746, 38980 - both of which I've
> closed today) show a history of incorrect URL-unescaping in mod_proxy.
Since then I've found several more duplicates in bugzilla.
oxy) must not change the URI.
This is exactly the bug I'm looking to fix.
> The reverse proxy (gateway) is just an origin server with a
> stupid name -- it must send a redirect if it makes the above
> change to a URI.
That would then be handled at the uri_decode stage, before
mo
ing the workers isn't
going to work, then managing them dynamically in a useful
manner is going to be a complex job that should probably
have its own separate module, based on a reslist and a
usage-counting strategy to drop little-used workers.
Or something like that.
--
Nick Kew
Application Dev
nd reverse proxy modules, a worker management module
will be a useful optional extra for both to have.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
On Tue, 18 Sep 2007 15:01:33 +0530
rahul <[EMAIL PROTECTED]> wrote:
> Hi,
> [Nick Kew:]
> ...
> | PR#35256 (which I was on the point of entering anew when I
> | found it). The simple patch to 35256 fixes the specific
> | instance of un-breaking AllowEncodedSlashes, but
k, requiring ISPs and hosting
> companies to log the originating port of all traffic.
Is there a reference for that legislation, and whatever debate there
was surrounding it? As in, what do they expect to gain from it?
> Any feedback is appreciated :)
Looks harmless, and evidently adds v
he proxy's address space.
Any interest?
--
Nick Kew
On Thu, 24 Jul 2003, Nick Kew wrote:
> 1. mod_diagnostics
Well, the feedback seems to indicate interest in that.
Shall I take that as a cue to do a writeup?
(BTW, someone said mod_bucketeer, but I can't see any overlap)
> 2. mod_upload
> 3. mod_xml_gnome_xslt
OK, no interest
od_upload/>
which serves to parse and decode multipart forms.
If you want to deal with URLencoded forms in a filter, that's easier:
just take some decoding code (e.g. from a CGI library, or write your own)
and put it in a filter.
--
Nick Kew
In urgent need of paying work - see http://www.w
gree with your points about its weaknesses.
What about that other inconsistency: the difference between
ap_rputs/family in Handlers and ap_fputs/etc in Filters? The subtle
difference to the arguments to these is - erm - annoying!
--
Nick Kew
les or usenet than dev.
--
Nick Kew
ssReverse should fix it.
I don't mind hacking mod_proxy to fix this if necessary, but I'd like
to check first that I'm not overlooking anything obvious.
If any PHP folks are listening, can I suggest PHP should prevent this
happening in the first place by fixing up Location: header
ed through.
(3) Where another filter's behavious is known to be safe
(mod_diagnostics filters can be removed at any time).
--
Nick Kew
d-proxy is very low traffic, FWIW. The reverse proxy was
expected to be low-traffic too, then unexpectedly jumped to 200K / day).
--
Nick Kew
odules for 2.0 that look at r->filename could call ap_request_filename() or
> ap_request_filename_str() depending on the requirement (dumb logging code would
> just call _str() version)
Even dumb logging code might be better with a printable placeholder,
yesno?
--
Nick Kew
On Thu, 30 Oct 2003, Nick Kew wrote:
> I've just hacked a couple of patches that fill gaps in Apache's
> URI manipulation, sufficient to construct HTTP requests for external
> resources. May I submit these for inclusion?
The only response to this was "where is it?"
-Encoding: gzip
Vary: Accept-Encoding
$ lynx -dump -source http://cvs.apache.org/snapshots/httpd-2.0-docs/ |wc
16 1301985
--
Nick Kew
that's where
his memory is going)?
I had a very similar situation to this, when a bug in a third-party
library caused it to buffer everything in my filter. mod_diagnostics
rapidly tracked that down for a x300 performance improvement.
--
Nick Kew
d
This does look a little like a request, but is not really sufficient.
> Examples:
This is useful thinking-it-through: what you're proposing looks a little
like a subrequest, and might ultimately make more sense. But it'll
need at least that move of the input filter chain!
--
Nick Kew
, we can use ap_provider on top of that to
declare and reference handlers.
--
Nick Kew
On Tuesday 04 October 2005 16:25, Brian Akins wrote:
> Nick Kew wrote:
> > The current implementation of handlers is a little bizarre, with each
> > handler both having to register itself and check its own name.
>
> Some people, myself included, have handlers that run even wh
inced we need another channel. We already have
#apache-modules, which deals with httpd (as well as module) hacking,
and has largely the same membership as #apr (and modest traffic -
we don't generally have a problem keeping helpdesk traffic to #apache:-)
--
Nick Kew
e know what's causing this?
The compiler is gcc-3.3.6.
[1] http://apache.webthing.com/mod_validator/
--
Nick Kew
V? I can see
a danger of some nasty bugs arising from confusion between this and
EOS buckets:
- Existing filters test for EOS and will ignore EOR. That's presumably
fine with you.
- Sooner or later, someone will write a filter that propagates EOR
and not EOS. Bang!
Is there no way you could use EOS directly?
--
Nick Kew
On Sunday 09 October 2005 18:49, Brian Pane wrote:
> On Oct 9, 2005, at 3:25 AM, Nick Kew wrote:
> > On Sunday 09 October 2005 02:37, [EMAIL PROTECTED] wrote:
> >> URL: http://svn.apache.org/viewcvs?rev=307339&view=rev
> >> Log:
> >> Redesign of reques
API rule:
Indeed. Better now than have two slightly-incompatible interpretations,
as we had with the "who owns a brigade" issue:-)
Problem closed, as far as I'm concerned - thanks:-)
--
Nick Kew
ere's some nice free googlerank for
http://dodgy.pills.example.com/?refid=yourstruly
--
Nick Kew
gh to support our own modules outside
the apache core distribution.
Talking of windows builds, "where's mod_ssl" seems to be something of
a FAQ in user support. Why is it a problem *now* to include it?
--
Nick Kew
t; for infrastructure@
But it's not just that. If we have links from the messages, that in itself
attracts the spammers! No matter whether we keep spam completely
out by moderating the lists, do cleanup after the event, or do nothing,
we should avoid offering major incentives to spammers.
--
Nick Kew
bly means something in the installer
that tells them they need openssl, and where to get it if necessary.
--
Nick Kew
around
that to-do list ... dammit, it's the long-awaited updates to charset_lite!
The harder bit to deal with is _local_ encoding in a different charsets in
header lines. That's a PITA, and is AFAIK peculiar to SMTP.
--
Nick Kew
int,
see Line 352 of util_filter.h.
--
Nick Kew
AuthDBDUserPWQuery "select password from authn where username = %s"
Require User nick
--
Nick Kew
/* Copyright 2000-2005 The Apache Software Foundation or its licensors, as
* applicable.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not
has on
HTTP headers rather than take on the full burden of fixing them itself.
--
Nick Kew
the code to allow unconditional use of 1.2.x features)
+1
Indeed, given not least the close relationship of the projects, I see no
real downside in requiring APR to be latest-at-time-of-2.2-release,
whether that be 1.2 or some later version.
--
Nick Kew
mod_authn_dbd seemed to get the thumbs-up from those who test-drove
it from trunk. Any thoughts on backporting to 2.2-branch at this stage?
It'll serve as a small and simple demo of using DBD, as well as in its
primary purpose.
--
Nick Kew
On Monday 24 October 2005 14:59, Mads Toftum wrote:
> On Mon, Oct 24, 2005 at 02:27:36PM +0100, Nick Kew wrote:
> > mod_authn_dbd seemed to get the thumbs-up from those who test-drove
> > it from trunk. Any thoughts on backporting to 2.2-branch at this stage?
> > It'll s
arity.
Any thoughts?
--
Nick Kew
e
AuthOrder user group dbm-group
which specifies an order of authz checks, and makes the last one
'authoritative' in terms of the old logic.
Of course that still leaves file-group looking lonely. Maybe what that
wants is a provider from authz_[file|dbm|dbd|ldap|etc] ?
--
Nick Kew
ir_config,
> &ep_auth_module);
The per-dir config is whatever you set it to. If it's null then presumably it
hasn't been set in this request. See http://www.apachetutor.org/dev/config
--
Nick Kew
d_dbd for that?
Your application is nontrivial but wasn't an input to the original
DBD design. That makes it exactly the kind of thing that'll prove
the API and/or highlight what needs adding or reviewing.
--
Nick Kew
Should we have an official attic for dead MPMs?
--
Nick Kew
rominently at the top?
OK, I'm ignoring Platform notes (I know nothing about non-unix)
and Other Topics (too big a subject for this post).
--
Nick Kew
ut that can't cross threads. When was that major
breakage discussed here?
--
Nick Kew
On Sunday 30 October 2005 23:07, Brian Pane wrote:
> On Oct 30, 2005, at 2:46 PM, Nick Kew wrote:
> > On Sunday 30 October 2005 22:40, Brian Pane wrote:
> >> Strictly speaking, there's no guarantee that a request will be
> >> processed by one and only one thread. I
option to tweak the behavior of content, once cached.
You mean as a tool for sysops to accept/decline serving from cache?
That could potentially have merit, and would work best in a quick-translation
hook to bypass any more complex/expensive rules.
The danger is if it grows some nightmarishly confusing relationship
to normal semantics: the existing vs
is bad enough for non-expert users!
--
Nick Kew
301 - 400 of 1350 matches
Mail list logo
