Re: [PATCH]DOC: LUA: fix some typos and syntax errors

2016-02-16 Thread Godbach
Hi Willy, On 2/16/16 7:48 PM, Willy Tarreau wrote: On Tue, Feb 16, 2016 at 07:15:07PM +0800, Godbach wrote: BTW, I have noticed that only master was applied by this patch. 1.6 branch should be also applied in my opinion. Yes it will definitely be, I'm lagging behind quite a bit

Re: [PATCH]DOC: LUA: fix some typos and syntax errors

2016-02-16 Thread Godbach
Hi Willy, On 16/2/16 6:10PM, Willy Tarreau wrote: Hi Godbach, On Tue, Feb 16, 2016 at 12:19:09PM +0800, Godbach wrote: Hi Willy, I have reviewed the DOC lua.txt twice. Attached is a patch for this documentation to fix some typos and syntax errors. The patch should be also backported to 1.6

[PATCH]DOC: LUA: fix some typos and syntax errors

2016-02-15 Thread Godbach
Hi Willy, I have reviewed the DOC lua.txt twice. Attached is a patch for this documentation to fix some typos and syntax errors. The patch should be also backported to 1.6. Any further amendment for this patch will be much appreciated. -- Best Regards, Godbach From

Re: [ANNOUNCE] haproxy-1.6.0 now released!

2015-10-16 Thread Godbach
Greate. A lot of new features and optimizations! -- Best Regards, Godbach

Re: [PATCH] remove codes for cleaning p-block_rules

2015-06-09 Thread Godbach
to update 1.6 branch(maybe caused by my local network). As you said, both 1.6 and 1.5 branches should apply this patch. -- Best Regards, Godbach From f8fa9c908b5b817e1a5804584bc8433ab91f4767 Mon Sep 17 00:00:00 2001 From: Godbach nylzhao...@gmail.com Date: Tue, 9 Jun 2015 19:41:52 +0800 Subject

[PATCH] remove codes for cleaning p-block_rules

2015-06-08 Thread Godbach
, Godbach

[PATCH] BUG/MINOR: parse: check the validity of size string in a more strict way

2015-01-28 Thread Godbach
it is necessary to apply this patch, both 1.6 and 1.5 need it. -- Best Regards, Godbach From 174943fb20fb3b45f186a6536b53151bdf00fee7 Mon Sep 17 00:00:00 2001 From: Godbach nylzhao...@gmail.com Date: Wed, 28 Jan 2015 17:36:16 +0800 Subject: [PATCH] BUG/MINOR: parse: check the validity of size

Re: [PATCH] BUG/MINOR: parse: check the validity of size string in a more strict way

2015-01-28 Thread Godbach
Hi Willy, On 2015/1/28 18:28, Willy Tarreau wrote: Hi Godbach, On Wed, Jan 28, 2015 at 05:57:13PM +0800, Godbach wrote: Hi Willy, Attached is a patch for parse_size_err(). If a stick table is defined as below: stick-table type ip size 50ka expire 300s HAProxy will stop parsing size

Re: [ANNOUNCE] haproxy-1.5.10

2015-01-03 Thread Godbach
the connection layer). In the mean time, I wish you all a happy new year! Willy Great work! Happy new year! -- Best Regards, Godbach

[BUG Report] BUG/MINOR: parse: refer curproxy instead of proxy

2014-12-18 Thread Godbach
this change in latest snapshot in your opinion? I can send a patch when I get this work done if you will accept the change. -- Best Regards, Godbach From c0f268d2638c0053fb470cdeda5cd70ed5cc0b96 Mon Sep 17 00:00:00 2001 From: Godbach nylzhao...@gmail.com Date: Thu, 18 Dec 2014 15:44:58 +0800 Subject

Re: [BUG Report] Maybe a bug for epoll

2014-12-17 Thread Godbach
of absmaxevents In the end, it seems that marking the bug as BUG/MINOR or OPTIM/MINOR is OK. Since some memory will be saved by the fix, I'd prefer to OPTIM/MINOR if you don't mind. -- Best Regards, Godbach

Re: [BUG Report] Maybe a bug for epoll

2014-12-17 Thread Godbach
Hi Willy, On 2014/12/17 16:03, Godbach wrote: On 2014/12/17 15:15, Willy Tarreau wrote: In my opinion, absmaxevents should be passed to epoll_wait() instead of global.tune.maxpollevents. No, that would defeat the principle of maxpollevents. In fact, I found what happened : commit f2e8ee2b

Re: [BUG Report] Maybe a bug for epoll

2014-12-17 Thread Godbach
Hi Willy, On 2014/12/17 16:40, Willy Tarreau wrote: In my opinion it's a cleanup since we're removing a leftover from an ancient piece of code :-) Willy OK, I agree with you and please modify the patch directly. -- Best Regards, Godbach

Re: [BUG Report] Maybe a bug for epoll

2014-12-17 Thread Godbach
Hi Willy, On 2014/12/18 0:07, Willy Tarreau wrote: Hi Godbach, On Wed, Dec 17, 2014 at 04:41:33PM +0800, Godbach wrote: From 798d50b1922e9680ea69cbc39d54f20665c782da Mon Sep 17 00:00:00 2001 From: Godbach nylzhao...@gmail.com Date: Wed, 17 Dec 2014 16:14:26 +0800 Subject: [PATCH 1/2] OPTIM

[BUG Report] Maybe a bug for epoll

2014-12-16 Thread Godbach
= epoll_wait(epoll_fd, epoll_events, absmaxevents, wait_time); tv_update_date(wait_time, status); measure_idle(); If the fix is right, I will send a patch later. -- Best Regards, Godbach

[PATCH] DOC: fix typo

2014-12-09 Thread Godbach
the patch for detailed information. In addition, this patch should be backported to 1.5 if necessary. -- Best Regards, Godbach From 73f3617908e1da775f10a2116394a1654d150d0f Mon Sep 17 00:00:00 2001 From: Godbach nylzhao...@gmail.com Date: Wed, 10 Dec 2014 10:21:30 +0800 Subject: [PATCH] DOC: fix typo

Re: haproxy 1.5.8 segfault

2014-11-21 Thread Godbach
Hi Willy, On 2014/11/21 14:35, Willy Tarreau wrote: Hi Godbach! On Fri, Nov 21, 2014 at 11:02:52AM +0800, Godbach wrote: Hi Willy, On 2014/11/19 2:31, Willy Tarreau wrote: On Tue, Nov 18, 2014 at 08:23:57PM +0200, Denys Fedoryshchenko wrote: Thanks! Seems working for me :) Will test more

Re: haproxy 1.5.8 segfault

2014-11-20 Thread Godbach
which allows user to enable or disable memset() in pool_alloc2() can be added into HAProxy. Since it is not an urgent issue, just take your time. :-) -- Best Regards, Godbach

Re: How to ignore backend busy server

2014-11-10 Thread Godbach
number of connections receives the new connection: http://cbonte.github.io/haproxy-dconv/configuration-1.5.html#method -- Best Regards, Godbach

[BUG report] haproxy will crash with some rewrite operation

2014-10-30 Thread Godbach
the end of the buffer */ memmove(pos + delta, pos, bi_end(b) - pos); This patch can be work well for both 1.5 release and 1.6-dev. Please help review this patch and I will send a patch if it is OK. -- Best Regards, Godbach

Re: [BUG report] haproxy will crash with some rewrite operation

2014-10-30 Thread Godbach
Hi Willy, On 2014/10/30 21:48, Willy Tarreau wrote: Hi Godbach, [ and first, sorry for not having yet responded to your other mail about caching ] Take it easy. :) Yes, it looks fine to me, feel free to send a patch and tag it as BUG/MAJOR. In the future we'll probably need to rework

Re: Question about srv_conn and beconn

2014-10-23 Thread Godbach
Hi, On 2014/10/23 13:24, 王继红 wrote: Hi Godbach Another question I want to ask, I see when the backend web server response time is more than 50s, then the haproxy will return http 504 to the client user, if the web server response the request but the response content is partly

Re: Question about srv_conn and beconn

2014-10-19 Thread Godbach
that there is an active connection on backend but has not been assigned to any server. -- Best Regards, Godbach

[PATCH] BUG/MINOR: server: move the directive #endif to the end of file

2014-07-28 Thread Godbach
as server_throttle_rate(), server_is_draining(), srv_adm_set_maint() and so on. Just move #endif directive to the end of file to solve this issue. -- Best Regards, Godbach From 4639eb27ae2dbe4cd047dcf30c51f9190195b1b5 Mon Sep 17 00:00:00 2001 From: Godbach nylzhao...@gmail.com Date: Mon, 28 Jul 2014 17:31:57

How to rewirte both URI and Host

2014-05-13 Thread Godbach
to rewrite both Host and URI reqrep expresion for rewritting host if match_host_uri reqrep expresion for rewritting URI if match_host_uri Since Host has been rewritten in firest reqrep directive, as I know, the second reqrep will not be executed as the match will fail. -- Best Regards, Godbach

Re: How to rewirte both URI and Host

2014-05-13 Thread Godbach
On 2014/5/13 15:05, Willy Tarreau wrote: Hi Godbach, On Tue, May 13, 2014 at 03:01:58PM +0800, Godbach wrote: Hi Willy, If I want to rewrite both Host and URI in request, and they are also in the match conditions, can HAProxy implement this requirement? The ACLs are as below: acl

Re: How to rewirte both URI and Host

2014-05-13 Thread Godbach
On 2014/5/13 16:05, Godbach wrote: On 2014/5/13 15:05, Willy Tarreau wrote: Hi Godbach, On Tue, May 13, 2014 at 03:01:58PM +0800, Godbach wrote: Hi Willy, If I want to rewrite both Host and URI in request, and they are also in the match conditions, can HAProxy implement this requirement

Re: [ANNOUNCE] haproxy-1.5-dev23

2014-04-23 Thread Godbach
misplaced http-request rules - MEDIUM: config: report misplaced use-server rules - DOC: update roadmap with what was done. --- Hi Willy, Great job. Thank you. -- Best Regards, Godbach

[PATCH] two minor patches about typo and code style

2014-04-21 Thread Godbach
Hi Willy, Please check the atthachments for two minor patches about typo and code style fix. -- Best Regards, Godbach From 3013eada648ee4134c27e0a38d75e4cc4e5f97d9 Mon Sep 17 00:00:00 2001 From: Godbach nylzhao...@gmail.com Date: Mon, 21 Apr 2014 21:42:41 +0800 Subject: [PATCH 1/2] DOC: fix

Re: Changing server weights without stopping the service

2014-03-24 Thread Godbach
that changing a server's weight on the fly will have no effect. -- Best Regards, Godbach

Re: RES: Bytes In Per Second

2014-03-02 Thread Godbach
. The configuration from the latest docmument is as below: rate-limit sessions rate Set a limit on the number of new sessions accepted per second on a frontend It seems that rate limit for sessions has not been implemented in backend. -- Best Regards, Godbach

Re: [ANNOUNCE] haproxy-1.5-dev21

2013-12-16 Thread Godbach
.html Thanks to all the early testers who reported issues, configs etc. Willy Great news! 1.5-dev21 may be the fastest one of all releases. :-) -- Best Regards, Godbach

[PATCH] two patches about code style and manual

2013-12-11 Thread Godbach
submit a new patch for this issue if you agree with me about this fix. -- Best Regards, Godbach From 35942e286901a50dad02a3cf355f0d943a137eec Mon Sep 17 00:00:00 2001 From: Godbach nylzhao...@gmail.com Date: Wed, 11 Dec 2013 19:48:57 +0800 Subject: [PATCH 1/2] MINOR: code style: use tabs to indent

Re: [PATCH] two patches about code style and manual

2013-12-11 Thread Godbach
Hi Willy, On 2013/12/11 20:43, Willy Tarreau wrote: Hi Godbach, There is also another problem I want to talk about stick table. The following codes are from lastest snapshot: (...) If stktable_init() fails, HAProxy will exit eventually. I can submit a new patch for this issue if you agree

Re: [PATCH] two patches about manual and stream_interface

2013-12-04 Thread Godbach
Hi Willy, On 2013/12/4 15:02, Willy Tarreau wrote: Hi Godbach, On Wed, Dec 04, 2013 at 11:50:36AM +0800, Godbach wrote: Hi Willy, Please check the attachments for the two patches. Patch 0001: DOC: stick-table can be declared in such sections as frontend, listen and backend OK in principle

Re: [PATCH] two patches about manual and stream_interface

2013-12-04 Thread Godbach
about this in the first mail but I mean we should have si_conn_send() return void so that the caller only checks ther flags, hence the removal of the comment above. Willy Got it. :-) Please check the attachment for the lastest patch. -- Best Regards, Godbach From

[PATCH] two patches about manual and stream_interface

2013-12-03 Thread Godbach
into patch0002 and send the patch again. -- Best Regards, Godbach From 1d9b14eadbfe88f99831e091ee2be6836b44a004 Mon Sep 17 00:00:00 2001 From: Godbach nylzhao...@gmail.com Date: Wed, 4 Dec 2013 10:14:50 +0800 Subject: [PATCH 1/2] DOC: stick-table can be declared in such sections as frontend, listen

Re: How do sticky tables work?

2013-11-14 Thread Godbach
-- Best Regards, Godbach

[PATCH] OPTIM: buffer: align the last output line of buffer_dump()

2013-11-13 Thread Godbach
Hi Willy, buffer_dump() in src/buffer.c is a usefull function to dump buffer to do debugging. However, the last output line should be aligned in a more readable way. Please check the attachment for the patch. -- Best Regards, Godbach From 7d19e9f3d35be3fb253ab586041b67312a03232f Mon Sep 17 00

Re: Complie Failed with DEBUG option -DDEBUG_FSM

2013-10-15 Thread Godbach
On 2013/10/15 1:15, Willy Tarreau wrote: Hi godbach, It would make more sense to completely redefine the output based on the information we have now in my opinion. So better dump the elements of the channel one by one instead of trying to map them to the old ones. Indeed, the only reason

Complie Failed with DEBUG option -DDEBUG_FSM

2013-10-14 Thread Godbach
by o=%p, what should it be. If it should be s-req-buf-p, it is the same as bi_ptr(s-req-buf) actually. If the replacements are correct and the use of o=%p can be confirmed, I will send a patch later. -- Best Regards, Godbach

Re: About si_conn_send_loop()

2013-10-11 Thread Godbach
On 2013/10/11 14:10, Willy Tarreau wrote: Hi Godbach, Yes I remember about this change, it happened one year ago in the following commit : ed7f836 BUG/MINOR: stream_interface: don't loop over -snd_buf() In 1.4 and 1.5 before the connection rework, the loop was used to call send() over each

About si_conn_send_loop()

2013-10-09 Thread Godbach
there is a 'break' with no condition in the last line of while loop, the loop will be only excuted once. It just looks like a 'if' as below: - while (!(conn-flags (CO_FL_ERROR | CO_FL_SOCK_WR_SH | ...))) { + if (!(conn-flags (CO_FL_ERROR | CO_FL_SOCK_WR_SH | ...))) { -- Best Regards, Godbach

Re: [PATCH] BUG/MINOR: deinit: free server map which is allocated in init_server_map()

2013-10-07 Thread Godbach
Hi Willy, On 2013-10-6 19:32, Willy Tarreau wrote: Hi Godbach, OK. Just for your info, the if before the free() is not needed because free() already does it, so I'll remove it. diff --git a/src/haproxy.c b/src/haproxy.c index ec9f513..265c39c 100644 --- a/src/haproxy.c +++ b/src/haproxy.c

Re: [PATCH] BUG/MINOR: deinit: free server map which is allocated in init_server_map()

2013-10-07 Thread Godbach
On 2013-10-7 15:47, Willy Tarreau wrote: Hi Godbach, On Mon, Oct 07, 2013 at 03:37:19PM +0800, Godbach wrote: Yes, I have checked the manual of free(), it said that If ptr is NULL, no operation is performed. In fact it's true on *most* platforms, and all those supported by haproxy. So we

[PATCH] BUG/MINOR: deinit: free server map which is allocated in init_server_map()

2013-10-02 Thread Godbach
if one of the above load balance algorithms is used. -- Best Regards, Godbach From 9366d553d9415c3d3d7f6889903eca2e7382f597 Mon Sep 17 00:00:00 2001 From: Godbach nylzhao...@gmail.com Date: Wed, 2 Oct 2013 17:10:11 +0800 Subject: [PATCH] BUG/MINOR: deinit: free server map which is allocated

Re: [PATCH] DOC: fix typo in comments

2013-10-01 Thread Godbach
On 2013-10-1 15:49, Willy Tarreau wrote: Hi Godbach, On Mon, Sep 30, 2013 at 11:23:10AM +0800, Godbach wrote: Hi Willy, There is a patch to fix typo in comments, please check the attachment for you information. The commit log is as below: commit 9824d1b3740ac2746894f1aa611c795366c84210

Re: [PATCH] DOC: fix typo in comments

2013-10-01 Thread Godbach
On 2013-10-2 1:12, Willy Tarreau wrote: On Wed, Oct 02, 2013 at 01:08:50AM +0800, Godbach wrote: It seems that you submit the whole text of mail as commit log to the git repository. Ah yes, you're right :-) At the moment I applied it, I was fooled by the subject into thinking it was just

[PATCH] DOC: fix typo in comments

2013-09-29 Thread Godbach
Hi Willy, There is a patch to fix typo in comments, please check the attachment for you information. The commit log is as below: commit 9824d1b3740ac2746894f1aa611c795366c84210 Author: Godbach nylzhao...@gmail.com Date: Mon Sep 30 11:05:42 2013 +0800 DOC: fix typo in comments

Re: Does Haproxy can deal with any request for one long conntect?

2013-08-15 Thread Godbach
request from the same connection, you should enable option http-server-close. As far as I know, haproxy maybe can not fulfill your requirement. -- Best Regards, Godbach

Re: Attempting to clear entries in stick table based on server id, results in all entries being dropped.

2013-08-14 Thread Godbach
On 2013/8/14 13:10, Willy Tarreau wrote: Hi Godbach, On Wed, Aug 14, 2013 at 10:20:10AM +0800, Godbach wrote: I have done a new test with this patch, it works well now. Thanks for testing. Yes, just do the same test as last node for other nodes to be removed. I had tried to fixed it just

Re: Attempting to clear entries in stick table based on server id, results in all entries being dropped.

2013-08-13 Thread Godbach
On 2013/8/13 22:54, Willy Tarreau wrote: Hi guys, On Thu, Aug 08, 2013 at 08:48:53PM +0800, Godbach wrote: On 2013/8/8 18:50, Mark Brooks wrote: The issue I am seeing is that using the dev version of HAProxy 1.5-dev19 git commit id 00f0084752eab236af80e61291d672e835790cff I have a source IP

Re: Attempting to clear entries in stick table based on server id, results in all entries being dropped.

2013-08-08 Thread Godbach
but a pity that I cannot fix it completely by now. :-( -- Best Regards, Godbach

Re: keep-alive connections to the backend

2013-08-08 Thread Godbach
in a first time (just check the target). HAProxy will release 1.5 stable version with this feature. -- Best Regards, Godbach

Re: Different check conditions for server selected from lb trees

2013-08-06 Thread Godbach
On 2013/8/1 17:05, Godbach wrote: Hi Willy, Haproxy will check the number of concurrent sessions assigned to the server after being selected from lb trees. Both leastconn and roundrobin use the same condition as below: fwlc_get_next_server(): if (!s-maxconn || (!s-nbpend s-served

Re: Different check conditions for server selected from lb trees

2013-08-06 Thread Godbach
On 2013/8/6 20:36, Willy Tarreau wrote: Hi Godbach, Sorry for replying late, I don't have a regular internet access at the moment. On Tue, Aug 06, 2013 at 07:48:46PM +0800, Godbach wrote: It seems that static-rr should also use the same check condition for the server after being selected

Different check conditions for server selected from lb trees

2013-08-01 Thread Godbach
Regards, Godbach

Re: Confused by the behaviour of calloc during init() in haproxy.c

2013-07-24 Thread Godbach
0:00 ./haproxy -f h.cfg -d The RSS is 113Mbytes with maxconn is 1,048,576. If I decrease the maxconn to be half, RSS is 55Mbytes. It seems that dev7 is somehow different from lastest snapshot in the same condition. -- Best Regards, Godbach

Re: Confused by the behaviour of calloc during init() in haproxy.c

2013-07-24 Thread Godbach
On 2013/7/24 16:18, Willy Tarreau wrote: On Wed, Jul 24, 2013 at 03:55:23PM +0800, Godbach wrote: On 2013/7/24 13:42, Willy Tarreau wrote: I agree and can confirm your results here. Also, if I replace the calloc with a malloc, I still see mmap and we don't have the 300M RSS anymore : willy

Re: Confused by the behaviour of calloc during init() in haproxy.c

2013-07-24 Thread Godbach
On 2013/7/24 17:07, Willy Tarreau wrote: On Wed, Jul 24, 2013 at 04:58:31PM +0800, Godbach wrote: The mechanism of poisoning/memset is used for memory pool. Memory pool uses MALLOC in dev7 and CALLOC in master, most chunks of memory are allocated from memory pool while processing sessions

Re: Roundrobin can not work well when a server's weight is 256

2013-07-23 Thread Godbach
2013/7/21 Godbach nylzhao...@gmail.com On 2013-7-21 21:17, Willy Tarreau wrote: No, I really mean UWGHT_MAX set to UWGHT_RANGE since it is used to fix set highest user weight that can be set. Willy Yeah, I get what you meant now. With the following fix: diff --git a/include/types

Re: Roundrobin can not work well when a server's weight is 256

2013-07-23 Thread Godbach
2013/7/22 Willy Tarreau w...@1wt.eu On Sun, Jul 21, 2013 at 10:56:26PM +0800, Godbach wrote: On 2013-7-21 16:53, Willy Tarreau wrote: Hi Godbach, I carefully checked and for me we can set SRV_UWGHT_MAX to SRV_UWGHT_RANGE. It will have the effect of limiting the maximum number

Re: Roundrobin can not work well when a server's weight is 256

2013-07-21 Thread Godbach
On 2013-7-21 16:53, Willy Tarreau wrote: Hi Godbach, That's quite a good analysis. I'm realizing that durnig the development of the algorithm, the maximum weight was 255, and it was later changed to 256. The test code that was used for this is still in tests/filltab25.c if you're curious

Re: Roundrobin can not work well when a server's weight is 256

2013-07-21 Thread Godbach
On 2013-7-21 18:16, Godbach wrote: On 2013-7-21 16:53, Willy Tarreau wrote: Hi Godbach, That's quite a good analysis. I'm realizing that durnig the development of the algorithm, the maximum weight was 255, and it was later changed to 256. The test code that was used for this is still

Re: Roundrobin can not work well when a server's weight is 256

2013-07-21 Thread Godbach
together. I will finish this job carefully and post patches ASAP. -- Best Regards, Godbach

Re: Roundrobin can not work well when a server's weight is 256

2013-07-21 Thread Godbach
On 2013-7-21 16:53, Willy Tarreau wrote: Hi Godbach, I carefully checked and for me we can set SRV_UWGHT_MAX to SRV_UWGHT_RANGE. It will have the effect of limiting the maximum number of servers a full weight to 4095 instead of 4128 (so please update the doc for this). It is also used

Roundrobin can not work well when a server's weight is 256

2013-07-20 Thread Godbach
)) { int w; w = atol(args[cur_arg + 1]); if (w 0 || w 256) { If so, I will give a patch to fix it, or it should be fixed in other way. -- Best Regards, Godbach

Re: Confused by the behaviour of calloc during init() in haproxy.c

2013-07-19 Thread Godbach
On 2013/7/19 14:34, Willy Tarreau wrote: On Fri, Jul 19, 2013 at 01:55:43PM +0800, Godbach wrote: Hi Willy, Here is part information about strace during haproxy startup: open(h.cfg, O_RDONLY) = 3 fstat(3, {st_mode=S_IFREG|0644, st_size=1985, ...}) = 0 mmap(NULL, 4096

Re: Confused by the behaviour of calloc during init() in haproxy.c

2013-07-18 Thread Godbach
Hi Willy, On 2013/7/19 13:32, Willy Tarreau wrote: Hi Godbach, On Fri, Jul 19, 2013 at 12:21:23PM +0800, Godbach wrote: You should strace it. I'm sure you'll see an mmap() call that matches the size of your allocation, meaning that the libc has decided to use this instead of sbrk

Re: SSL Handshake errors

2013-07-08 Thread Godbach
with tcpdump -qns 0 -X -r file.cap translated.cap in order to make the dump readable and extract the two requests. If the original dump is needed, let me know and I'll attach it a.s.a.p. Hi, Andrei You'd better supply the original pcap file instead, not the readable format. -- Best Regards, Godbach

Re: Performance of concurrent connections decreases from 1.5-dev7 to 1.5-dev17

2013-07-06 Thread Godbach
On 2013/7/6 16:52, Willy Tarreau wrote: On Sat, Jul 06, 2013 at 04:31:49PM +0800, Godbach wrote: I've got what you meant. Once allocated from system, haproxy will not release the memory until it receives SIGQUIT or does soft stop. If less memory during running is expected, just decrease

Performance of concurrent connections decreases from 1.5-dev7 to 1.5-dev17

2013-07-05 Thread Godbach
hash-type consistent srv1... srv2... [40 servers] Best Regards, Godbach

[patch] Two minor patches about buffer and typo in DOC

2013-07-01 Thread Godbach
; } In my opinion, buffer_not_empty() returns a bool type, and logical OR is a better choiceas follow: static inline int buffer_not_empty(const struct buffer *buf) { return buf-i || buf-o; } Best Regards, Godbach From ecfcee76b3ac4db843dc3c85f168a2c789a4c653 Mon Sep 17 00:00:00 2001 From: Godbach

Re: [patch] Two minor patches about buffer and typo in DOC

2013-07-01 Thread Godbach
depends in fact. Anyway the first one is always better since it does not have to build new values. Best regards, Willy Hi, Willy Thanks for your detailed analysis. Cool! I have observed some optimizations in HAProxy before, it is a new one. Best Regards, Godbach -- Best Regards, Godbach

[PATCH] BUG/MINOR: deinit: free fdinfo while doing cleanup

2013-06-26 Thread Godbach
is starting, but only fdtab was freed in deinit(), fdinfo should also be freed. Best Regards, Godbach From b0c51207cb74ec01b3cdb0f79b3b60f359912a4a Mon Sep 17 00:00:00 2001 From: Godbach nylzhao...@gmail.com Date: Wed, 26 Jun 2013 16:49:51 +0800 Subject: [PATCH] BUG/MINOR: deinit: free fdinfo while

Re: About half close and reset connection by client

2013-06-21 Thread Godbach
option nolinger now. -- Best Regards, Godbach

Re: 'SSL handshake failure' errors

2013-06-20 Thread Godbach
Hi Merton, It is a good way to capture the packets during SSL handshake by tcpdump or wireshark from your client to find out what error happens. I have used this method in debugging SSL feature in haproxy. FYI. Best Regards, Godbach On 2013/6/20 1:46, Merton Lister wrote: Thank you Lukas

About half close and reset connection by client

2013-06-20 Thread Godbach
am wondering that what I have analysed is correct or not. If correct, it should be necessary to care about the events on frontend connection even that haproxy has not got response from backend server. -- Best Regards, Godbach

Re: About half close and reset connection by client

2013-06-20 Thread Godbach
On 2013/6/20 21:31, Godbach wrote: On 2013/6/20 21:18, Willy Tarreau wrote: Hi Godbach, On Thu, Jun 20, 2013 at 09:07:57PM +0800, Godbach wrote: Hi, Willy I have noticed that half-closed timeout appears in ROADMAP file. That't the issue I have tested several days ago. I have done a test

[patch] CLEANUP: session: remove event_accept() which was not used anymore

2013-06-19 Thread Godbach
/proto_http.h and use correct function name in other two files instead of event_accept(). -- Best Regards, Godbach From 949b6e22353260b8d36178cb704bb88d4564894a Mon Sep 17 00:00:00 2001 From: Godbach nylzhao...@gmail.com Date: Thu, 20 Jun 2013 13:28:38 +0800 Subject: [PATCH] CLEANUP: session: remove

Re: HTTP Request still gets response from the server with weight 0 and src persistence

2013-05-29 Thread Godbach
size 200k expire 600s stick on src server 1 10.128.7.1:80 id 1 cookie srv1 weight 1 maxconn 0 slowstart 0s server 2 10.128.7.2:80 id 2 cookie srv2 weight 1 maxconn 0 slowstart 0s Best Regards, Godbach On 2013/5/29 13:35, Baptiste wrote: Hi Godbach

Re: HTTP Request still gets response from the server with weight 0 and src persistence

2013-05-29 Thread Godbach
On Wed, May 29, 2013 at 9:18 AM, Godbach nylzhao...@gmail.com wrote: Hi Baptiste Thanks for your replying. I am using the balance roundrobin algorithm and sticking on src, not the the balance source algorithm. The configuration has been presented in my first mail as below: backend pool

Re: HTTP Request still gets response from the server with weight 0 and src persistence

2013-05-29 Thread Godbach
Hi Baptiste Yeah, I got it. Thank you very much for your explanation. Best Regards, Godbach Actually, this is the purpose of dropping a weight to 0: being able to maintain sticky sessions. If you want to shutdown rudely your server, preventing everybody to access it, use the disable keyword

HTTP Request still gets response from the server with weight 0 and src persistence

2013-05-28 Thread Godbach
that whether the flag SRV_RUNNING of server should be cleared or not when its weight is set to 0. -- Best Regards, Godbach

Re: Limit frontend bandwidth rate?

2013-05-04 Thread Godbach
or not. After testing lastest snapshot, http keepalive has already token effect between haproxy and server under such conditions: * configuration without 'option http-server-close' set * server side has enabled http keepalive. Best Regards, Godbach

Re: Limit frontend bandwidth rate?

2013-05-04 Thread Godbach
of haproxy will be promoted. That's what I have understood, am I right? Best Regards, Godbach Hi Godbach! I am wondering that whether server side keepalive you said is http keepalive between haproxy and server or not. Thats it exactly. After testing lastest snapshot, http keepalive has

Re: Limit frontend bandwidth rate?

2013-05-04 Thread Godbach
backend connections and more data to be processed by userland in full keepalive mode, CPU shoudl be used more efficiently as you said. Best Regards, Godbach

Re: SMTP load balancer.

2013-05-01 Thread Godbach
balance roundrobin ... Then have a try. :-) Godbach

Re: SMTP load balancer.

2013-05-01 Thread Godbach
On 5/2/2013 5:46 AM, Godbach wrote: On 5/1/2013 8:39 AM, Bryan Talbot wrote: Looks like you've listed the same IP address twice. -Bryan Typo, Thanks. How have I missed that?? Eliezer Hi, Elizer You can use roundrobin instead of leastconn as below: listen tcp 0.0.0.0:225 mode

Re: Haproxy crash while initializing compression

2013-04-28 Thread Godbach
Hi Godbach, On Sun, Apr 28, 2013 at 12:16:17PM +0800, Godbach wrote: Hi, all Haproxy of latest snapshot will crash while initializing compression under such configuration: global ... tune.zlib.memlevel 6 tune.zlib.windowsize 8 frontend ... compression algo gzip

Re: CRL verification problem

2013-04-27 Thread Godbach
, X509_V_FLAG_CRL_CHECK|X509_V_FLAG_CRL_CHECK_ALL); I guess that it may be the behavior of SSL library but found nothing related with this problem. Any help will be much appreciated. Best Regards, Godbach The main configuration for frontend as below: frontend fe bind ip:443 ssl crt 1.cer verify

Haproxy crash while initializing compression

2013-04-27 Thread Godbach
order 'gzip deflate'. Zlib version is zlib-1.2.3. Best Regards, Godbach

CRL verification problem

2013-04-26 Thread Godbach
|X509_V_FLAG_CRL_CHECK_ALL); I guess that it may be the behavior of SSL library but found nothing related with this problem. Any help will be much appreciated. Best Regards, Godbach

Re: Failed to use the source address for outgoing connections

2013-04-23 Thread Godbach
Hi, (it's strange I didn't get the original e-mail). On Tue, Apr 23, 2013 at 6:27 AM, Godbach nylzhao...@gmail.com wrote: Hi, all I have tested 'source' config in haproxy-1.5-dev18, but it didn't work with the following line in default section: source 0.0.0.0 usesrc clientip Other

Re: Failed to use the source address for outgoing connections

2013-04-23 Thread Godbach
On Tue, Apr 23, 2013 at 02:44:17PM +0800, Godbach wrote: Hi, (it's strange I didn't get the original e-mail). On Tue, Apr 23, 2013 at 6:27 AM, Godbach nylzhao...@gmail.com wrote: Hi, all I have tested 'source' config in haproxy-1.5-dev18, but it didn't work with the following line

Re: Failed to use the source address for outgoing connections

2013-04-22 Thread Godbach
Hi, Yes. I have tried that and it works as expected. The original information in my last mail as below: Then I put source configuration 'source 0.0.0.0 usesrc clientip' in backend section, the source function can work well. Best Regards, Godbach Hi, Have you tried moving the source

Is there any plan to support OCSP to verify cert

2013-03-05 Thread Godbach
Hi, all OCSP(Online Certificate Status Protocol) is also used to verify certificates. I am wondering that if there is any plan to support OCSP in haproxy in the future. Best Regards, Godbach

Re: Is there any plan to support OCSP to verify cert

2013-03-05 Thread Godbach
certificate in the future. Best Regards, Godbach 2013/3/6 John Marrett jo...@zioncluster.ca: Godbach, I'm interested to better understand what you want to do with OSCP. Ordinarily if you present a certificate using haproxy clients will validate it using methods specified in the certificate

Re: Is there any plan to support OCSP to verify cert

2013-03-05 Thread Godbach
certificates in the future. Best Regards, Godbach 2013/3/6 John Marrett jo...@zioncluster.ca: Godbach, I'm interested to better understand what you want to do with OSCP. Ordinarily if you present a certificate using haproxy clients will validate it using methods specified in the certificate itself

Re: Failed to persist on SSL session ID

2013-02-26 Thread Godbach
Hi Bobby1 Maybe you forgot to set USE_OPENSSL=1 to in Makefile to support ssl. Godbach 2013/2/26 bobby1 b...@yahoo.com: Emeric BRUN ebrun@... writes: Ki, original message- De: Godbach nylzhaowei@... A: Emeric BRUN ebrun@... Copie à: haproxy

  1   2   >