once in a while I saw "reg-tests/compression/lua_validation.vtc" failed.
say 1 times out of 20.
it is slow and racy by nature.
also, seems 3 weeks of Linaro cloud are passed, 1 week is left.
I'll do more tries
ср, 11 мар. 2020 г. в 19:14, Martin Grigorov :
>
>
> On Wed, Mar 11, 2020 at 3:06 PM
сб, 14 мар. 2020 г. в 01:06, Willy Tarreau :
> On Fri, Mar 13, 2020 at 03:24:15PM +0100, William Lallemand wrote:
> > If you grep on BUILD: in the git log, this keyword does not mean
> > anything anymore. And this is confusing in my opinion. We could
> > introduce "CI: " instead.
>
> Good idea. I
On Fri, Mar 13, 2020 at 04:50:12PM +0500, ??? wrote:
> ??, 13 ???. 2020 ?. ? 16:48, Aleksandar Lazic :
>
> > Mar 13, 2020 12:11:16 PM ??? :
> >
> > > initial motivation was that article
> > >
> > > https://docs.travis-ci.com/user/multi-cpu-architectures
> >
> > Thanks for the lin
On Fri, Mar 13, 2020 at 11:18:42PM +0500, ??? wrote:
> Hello,
>
> there's always noise like that
>
>
> *** h1 debug|[WARNING] 072/163523 (6090) : config : log format
> ignored for proxy 'testme' since it has no log address.
> *** h1 debug|[WARNING] 072/163523 (6090) : config : log
On Fri, Mar 13, 2020 at 03:24:15PM +0100, William Lallemand wrote:
> If you grep on BUILD: in the git log, this keyword does not mean
> anything anymore. And this is confusing in my opinion. We could
> introduce "CI: " instead.
Good idea. I can adjust it.
Ilya, just a question, what will be the i
Hi Martin,
On Fri, Mar 13, 2020 at 12:35:12PM +0200, Martin Grigorov wrote:
> Hi ,
>
> Suddenly today the build is again green here!
> I didn't make any changes to my testing setup.
> It must be something on the OS level but I wasn't able to figure out what
> makes the HAProxy tests timeout i
Hello,
there's always noise like that
*** h1 debug|[WARNING] 072/163523 (6090) : config : log format
ignored for proxy 'testme' since it has no log address.
*** h1 debug|[WARNING] 072/163523 (6090) : config : log format
ignored for frontend 'test_abns' since it has no log address.
any sp
On Fri, Mar 13, 2020 at 05:41:14PM +0100, Tim Düsterhus wrote:
> Willy,
>
> Am 13.03.20 um 17:34 schrieb Willy Tarreau:
> > Indeed, just found it in my queue. However we usually use it
> > differently, with the function instead of the variable. Do you
> > mind if I adapt it ?
> >
>
> I attempted
Willy,
Am 13.03.20 um 17:34 schrieb Willy Tarreau:
> Indeed, just found it in my queue. However we usually use it
> differently, with the function instead of the variable. Do you
> mind if I adapt it ?
>
I attempted to use the function, but it didn't compile. I guess because
of a circular depend
On Fri, Mar 13, 2020 at 12:23:26PM +0100, Tim Düsterhus wrote:
> Willy,
>
> Am 09.03.20 um 17:05 schrieb Tim Duesterhus:
> > gcc complains (non-rightfully):
> >
> >> include/common/buf.h: In function 'br_head_pick':
> >> include/common/debug.h:62:4: warning: ignoring return value of 'write',
> >
On Fri, Mar 13, 2020 at 03:33:38PM +0100, Tim Düsterhus wrote:
> The v2-series is good from my side. I don't plan any more changes. If
> you are happy as well then please take it.
Perfect, series now applied, thanks!
Willy
Willy,
Am 13.03.20 um 14:18 schrieb Willy Tarreau:
> I was willing to take it as is but since I had a comment for the last
> one and I know that many times when reworking a patch we spot things
> we wish has been done differently in the previous ones, I let you
> decide what suits you best :-) If
On Fri, Mar 13, 2020 at 07:05:15PM +0500, Илья Шипицин wrote:
> Willy, can we apply this?
>
> I scheduled it on tuesday every week.
>
> вт, 10 мар. 2020 г. в 12:12, Илья Шипицин :
>
> > Hello,
> >
> > I implemented spell check CI.
> > it is a bit noisy, however, I hope we will polish it soon.
>
Willy, can we apply this?
I scheduled it on tuesday every week.
вт, 10 мар. 2020 г. в 12:12, Илья Шипицин :
> Hello,
>
> I implemented spell check CI.
> it is a bit noisy, however, I hope we will polish it soon.
>
> Cheers,
> Ilya Shipitcin
>
On Fri, Mar 13, 2020 at 11:57:14AM +0100, Tim Düsterhus wrote:
> Willy,
>
> Am 13.03.20 um 08:44 schrieb Willy Tarreau:
> >> I've added the new `struct connection` member at the end. Please check
> >> whether you think that is the appropriate place for it or if it should
> >> be moved somewhere el
пт, 13 мар. 2020 г. в 16:48, Aleksandar Lazic :
> Mar 13, 2020 12:11:16 PM Илья Шипицин :
>
> > initial motivation was that article
> >
> > https://docs.travis-ci.com/user/multi-cpu-architectures
>
> Thanks for the link.
>
> > travis allows to run builds on various archs, why not to test ?
>
> Ful
Mar 13, 2020 12:11:16 PM Илья Шипицин :
> initial motivation was that article
>
> https://docs.travis-ci.com/user/multi-cpu-architectures
Thanks for the link.
> travis allows to run builds on various archs, why not to test ?
Full Agreeing. ;-)
Would be interesting if anyone use it on Host or p
Willy,
squash this one if you like.
Best regards
Tim Düsterhus
Apply with `git am --scissors` to automatically cut the commit message.
-- >8 --
This adds the missing blank lines in `make_proxy_line_v2` and
`conn_recv_proxy`. It also adjusts the type of the temporary variable
used for the return
Willy,
here I removed the BUG_ON() and added the blank lines you requested.
> > + BUG_ON(sess != conn->owner); /* XXX is this correct? */
>
> I'm pretty sure this one will break from time to time, maybe during
> retries or when reusing an idle connection or something. Better drop
> it.
Willy,
this one has not been modified compared to v1.
Best regards
Tim Düsterhus
Apply with `git am --scissors` to automatically cut the commit message.
-- >8 --
This reserves and defines TLV type 0x05.
---
doc/proxy-protocol.txt | 20
include/types/connection.h | 1 +
Willy,
In this one I added a blank line after declaration that you missed during
review.
Best regards
Tim Düsterhus
Apply with `git am --scissors` to automatically cut the commit message.
-- >8 --
This patch reads a proxy protocol v2 provided unique ID and makes it
available using the `fc_pp_un
Willy,
Am 09.03.20 um 17:05 schrieb Tim Duesterhus:
> gcc complains (non-rightfully):
>
>> include/common/buf.h: In function ‘br_head_pick’:
>> include/common/debug.h:62:4: warning: ignoring return value of ‘write’,
>> declared with attribute warn_unused_result [-Wunused-result]
>> (void)wri
Willy,
Ilya,
Am 13.03.20 um 12:06 schrieb Willy Tarreau:
>> I just realized that BUG_ON is not active by default (should we add
>> DEBUG_STRICT=1 to CI?). The check in there does not even compile (facepalm).
>
> Yes, definitely a good idea.
I've filed an issue for the CI expert Ilya to look into
initial motivation was that article
https://docs.travis-ci.com/user/multi-cpu-architectures
travis allows to run builds on various archs, why not to test ?
пт, 13 мар. 2020 г. в 16:07, Aleksandar Lazic :
> Hi.
>
> I'm wondering that this target is tested.
>
> http://git.haproxy.org/?p=haproxy.gi
On Fri, Mar 06, 2020 at 01:15:40PM +0100, Tim Duesterhus wrote:
> @@ -1466,6 +1466,20 @@ int make_proxy_line_v2(char *buf, int buf_len, struct
> server *srv, struct connec
> ret += make_tlv(&buf[ret], (buf_len - ret),
> PP2_TYPE_AUTHORITY, value_len, value);
>
On Mon, Mar 09, 2020 at 05:07:38PM +0100, Tim Düsterhus wrote:
> Willy,
>
> Am 06.03.20 um 13:15 schrieb Tim Duesterhus:
> > is acceptable. There's also a `BUG_ON` left in the patch, because I wasn't
> > sure if I should grab the session from the stream or from the connection.
>
> I just realized
Hi.
I'm wondering that this target is tested.
http://git.haproxy.org/?p=haproxy.git;a=commitdiff;h=d726386421dcd184ca2518d17332f82e9cd79f2d
Are there really user which runs HAProxy on Host? 8-O
How perform HAProxy on that platform?
Regards
Aleks
Willy,
Am 13.03.20 um 08:44 schrieb Willy Tarreau:
>> I've added the new `struct connection` member at the end. Please check
>> whether you think that is the appropriate place for it or if it should
>> be moved somewhere else because of holes or caches.
>
> I'm seeing that the struct connection h
Hi Илья,
Suddenly today the build is again green here!
I didn't make any changes to my testing setup.
It must be something on the OS level but I wasn't able to figure out what
makes the HAProxy tests timeout in the last several days.
Regards,
Martin
On Wed, Mar 11, 2020 at 4:13 PM Martin Grigoro
On Fri, Mar 13, 2020 at 08:44:59AM +0100, Willy Tarreau wrote:
> I'm seeing that the struct connection has become huge (160 bytes now)
By the way, regarding this, I'm seeing this order of criticity in terms
of structures sizes:
1) struct fdtab : 64 bytes, cache-line-aligned, shared between threa
Hi Tim,
On Fri, Mar 06, 2020 at 01:15:39PM +0100, Tim Duesterhus wrote:
> For the whole unique ID processing I've used the new `ist` helpers, it
> might make sense to update the authority processing to make use of ist
> in a future CLEANUP patch.
>
> I've added the new `struct connection` member
Hi again
In version 2.1.3:
When config spoe engine proxys > 1, the function "srandom" will run more
than one times, it will make some engine_id duplicated, and the SPOA
is group SPOP connections by engine_id when option async is on, the
ack will reply to the wrong SPOE.
This patch should backport
32 matches
Mail list logo
