Withdrawn the proposal in r1747668 after reading the comment from Roy again.
Michael, since you know more about this: is there a specific UA string where
httpd can detect the broken NodeJS client from and suppress "Update: XXX"
response headers? I believe NodeJS is broken on *any* Update respons
Zitat von Stefan Eissing :
Withdrawn the proposal in r1747668 after reading the comment from Roy again.
Apache is the only HTTP/2 server in this world that sends this strange
header. So omitting it would be the right thing to do.
Michael, since you know more about this: is there a specific
Thx! Are any of these in trunk, or are they 2.4-branch specific?
> On Jun 9, 2016, at 1:03 PM, Jacob Champion wrote:
>
> Hi all,
>
> It may be too late now that the T&R machinery is in motion, but I'll bump
> some of my patchsets:
>
> - APXS: https://bz.apache.org/bugzilla/show_bug.cgi?id=589
This patch fixes an instance of undefined behavior in mod_http2 with
LogLevel >= trace2.
Please see the h2_h2_process_conn() function in h2_h2.c:631. The
call to ap_log_cerror() passes a pointer to a non-null terminated buffer
while specifying %s in the format string. This causes an out-of-bound
I noticed that currently mod_proxy_http2 cannot serve responses larger
than 32 KiB for incoming requests over HTTP/1.1. Below is a simple
config that can be used to reproduce the problem:
LoadModule http2_module modules/mod_http2.so
LoadModule proxy_module modules/mod_proxy.so
LoadMod
On Fri, Jun 10, 2016 at 4:03 AM, Stefan Eissing <
stefan.eiss...@greenbytes.de> wrote:
> Withdrawn the proposal in r1747668 after reading the comment from Roy
> again.
>
Let's look at what Roy said... that httpd's implementation of HTTP/1.1 may
choose
to honor a Connection-Upgrade request header
Thanks for the patch... It came thru, at least for me, mangled.
Could you resend?
> On Jun 10, 2016, at 10:21 AM, Evgeny Kotkov
> wrote:
>
>
Saw this today, and noticed they used httpd as one of the projects they
collected data about:
http://neverworkintheory.org/2016/06/09/too-many-knobs.html
http://cseweb.ucsd.edu/~tixu/papers/fse15.pdf
Hard to make a quick conclusion, but I do believe in the general statement
of enumerations > nume
Evgeny, the patch looks good. Thanks!
Will probably find the time early next week to get it in. Need to add some test
cases where the frontend connection is running HTTP/1.1. Although not the
primary use case, it needs to get tested. Thanks for looking into this.
-Stefan
> Am 10.06.2016 um 16:
Jim Jagielski writes:
> Thanks for the patch... It came thru, at least for me, mangled.
>
> Could you resend?
I think that's because gmail did base64 encoding for the attachment.
The unmangled patch is available here:
https://mail-archives.apache.org/mod_mbox/httpd-dev/201606.mbox/raw/%3CCAP_G
Thanks!
> On Jun 10, 2016, at 10:58 AM, Evgeny Kotkov
> wrote:
>
> Jim Jagielski writes:
>
>> Thanks for the patch... It came thru, at least for me, mangled.
>>
>> Could you resend?
>
> I think that's because gmail did base64 encoding for the attachment.
>
> The unmangled patch is availabl
I think we should add and submit for backport while we, in
parallel, add the tests... Seems like this should be in 2.4.21
which I hope to T&R next week.
> On Jun 10, 2016, at 10:53 AM, Stefan Eissing
> wrote:
>
> Evgeny, the patch looks good. Thanks!
>
> Will probably find the time early next
Ok, added in r1747735. The h2 test suite runs fine with it.
I am on the jump, if you like you should be able to just merge this back to
2.4.x.
-Stefan
> Am 10.06.2016 um 17:19 schrieb Jim Jagielski :
>
> I think we should add and submit for backport while we, in
> parallel, add the tests... S
On 06/10/2016 05:01 AM, Jim Jagielski wrote:
Thx! Are any of these in trunk, or are they 2.4-branch specific?
None of these are currently committed to trunk, but all the patches
should apply cleanly (or at least, they do for me, on a trunk from
sometime yesterday).
The majority of my person
On Tue, Jun 7, 2016 at 2:29 AM, Nick Gearls wrote:
> A global option is not a good option as it would disable this very useful
> check for all macros.
>
> What about "
15 matches
Mail list logo
