Willy Tarreau [Sun, 10 Jan 2010 13:21:19 +0000 (14:21 +0100)]
[MINOR] http: differentiate waiting for new request and waiting for a complete requst
While waiting in a keep-alive state for a request, we want to silently
close if we don't get anything. However if we get a partial request it's
different because that means the client has started to send something.
This requires a new transaction flag. It will be used to implement a
distinct timeout for keep-alive and requests.
Willy Tarreau [Sun, 10 Jan 2010 09:49:11 +0000 (10:49 +0100)]
[MINOR] http: move appsession 'sessid' from session to http_txn
This change, suggested by Cyril Bonté, makes a lot of sense and
would have made it obvious that sessid was not properly initialized
while switching to keep-alive. The code is now cleaner.
Willy Tarreau [Sun, 10 Jan 2010 09:21:21 +0000 (10:21 +0100)]
[BUG] stream_interface: fix retnclose and remove cond_close
The stream_int_cond_close() function was added to preserve the
contents of the response buffer because stream_int_retnclose()
was buggy. It flushed the response instead of flushing the
request. This caused issues with pipelined redirects followed
by error messages which ate the previous response.
This might even have caused object truncation on pipelined
requests followed by an error or by a server redirection.
Now that this is fixed, simply get rid of the now useless
function.
Cyril Bonté [Sat, 9 Jan 2010 23:30:14 +0000 (00:30 +0100)]
[BUG] appsession: possible memory leak in case of out of memory condition
I've tried to follow all the pool_alloc2/pool_free2 calls in the code
to track memory leaks. I've found one which only happens when there's
already no more memory when allocating a new appsession cookie.
Willy Tarreau [Sat, 9 Jan 2010 23:42:19 +0000 (00:42 +0100)]
[MINOR] http redirect: add the ability to append a '/' to the URL
Sometimes it can be desired to return a location which is the same
as the request with a slash appended when there was not one in the
request. A typical use of this is for sending a 301 so that people
don't reference links without the trailing slash. The name of the
new option is "append-slash" and it can be used on "redirect"
statements in prefix mode.
Willy Tarreau [Sat, 9 Jan 2010 23:24:22 +0000 (00:24 +0100)]
[MINOR] http: fix double slash prefix with server redirect
When using server redirection, it is possible to specify a path
consisting of only one slash. While this is discouraged (risk of
loop) it may sometimes be useful combined with content switching.
The prefixing of a '/' then causes two slashes to be returned in
the response. So we now do as with the other redirects, don't
prepend a slash if it's alone.
Willy Tarreau [Sat, 9 Jan 2010 23:15:35 +0000 (00:15 +0100)]
[MEDIUM] http: fix handling of message pointers
Some message pointers were not usable once the message reached the
HTTP_MSG_DONE state. This is the case for ->som which points to the
body because it is needed to parse chunks. There is one case where
we need the beginning of the message : server redirect. We have to
call http_get_path() after the request has been parsed. So we rely
on ->sol without counting on ->som. In order to achieve this, we're
making ->rq.{u,v} relative to the beginning of the message instead
of the buffer. That simplifies the code and makes it cleaner.
Willy Tarreau [Sat, 9 Jan 2010 20:29:23 +0000 (21:29 +0100)]
[BUG] server redirection used an uninitialized string.
This might have been introduced with chunk extensions. Note that
the server redirect still does not work because http_get_path()
cannot get the correct path once the request message is in the
HTTP_MSG_DONE state (->som does not point to the start of message
anymore).
Willy Tarreau [Fri, 8 Jan 2010 06:49:44 +0000 (07:49 +0100)]
[RELEASE] Released version 1.4-dev6
Released version 1.4-dev6 with the following main changes :
- [BUILD] warning in stream_interface.h
- [BUILD] warning ultoa_r returns char *
- [MINOR] hana: only report stats if it is enabled
- [MINOR] stats: add "a link" & "a href" for sockets
- [MINOR]: stats: add show-legends to report additional informations
- [MEDIUM] default-server support
- [BUG]: add 'observer', 'on-error', 'error-limit' to supported options list
- [MINOR] stats: add href to tracked server
- [BUG] stats: show UP/DOWN status also in tracking servers
- [DOC] Restore ability to search a keyword at the beginning of a line
- [BUG] stats: cookie should be reported under backend not under proxy
- [BUG] cfgparser/stats: fix error message
- [BUG] http: disable auto-closing during chunk analysis
- [BUG] http: fix hopefully last closing issue on data forwarding
- [DEBUG] add an http_silent_debug function to debug HTTP states
- [MAJOR] http: fix again the forward analysers
- [BUG] http_process_res_common() must not skip the forward analyser
- [BUG] http: some possible missed close remain in the forward chain
- [BUG] http: redirect needed to be updated after recent changes
- [BUG] http: don't set no-linger on response in case of forced close
- [MEDIUM] http: restore the original behaviour of option httpclose
- [TESTS] add a file to test various connection modes
- [BUG] http: check options before the connection header
- [MAJOR] session: fix the order by which the analysers are run
- [MEDIUM] session: also consider request analysers added during response
- [MEDIUM] http: make safer use of the DONT_READ and AUTO_CLOSE flags
- [BUG] http: memory leak with captures when using keep-alive
- [BUG] http: fix for capture memory leak was incorrect
- [MINOR] http redirect: use proper call to return last response
- [MEDIUM] http: wait for some flush of the response buffer before a new request
- [MEDIUM] session: limit the number of analyser loops
Willy Tarreau [Thu, 7 Jan 2010 23:32:27 +0000 (00:32 +0100)]
[MEDIUM] session: limit the number of analyser loops
The initial code's intention was to loop on the analysers as long
as an analyser is added by another one. [This code was wrong due to
the while(0) which breaks even on a continue statement, but the
initial intention must be changed too]. In fact we should limit the
number of times we loop on analysers in order to limit latency.
Using maxpollevents as a limit makes sense since this tunable is
used for the exact same purposes. We may add another tunable later
if that ever makes sense, so it's very unlikely.
Willy Tarreau [Thu, 7 Jan 2010 23:30:20 +0000 (00:30 +0100)]
[MEDIUM] http: wait for some flush of the response buffer before a new request
If we accept a new request and that request produces an immediate
response (error, redirect, ...), then we may fail to send it in
case of pipelined requests if the response buffer is full. To avoid
this, we check the availability of at least maxrewrite bytes in the
response buffer before accepting a new pipelined request.
Willy Tarreau [Thu, 7 Jan 2010 23:26:50 +0000 (00:26 +0100)]
[MINOR] http redirect: use proper call to return last response
During a redirect, we used to send the last chunk of response with
stream_int_cond_close(). But this is wrong in case of pipeline,
because if the response already contains something, this function
will refrain from touching the buffer. Use a concatenation function
instead.
Also, this call might still fail when the buffer is full, we need
a second fix to refrain from parsing an HTTP request as long as the
response buffer is full, otherwise we may not even be able to return
a pending redirect or an error code.
Willy Tarreau [Thu, 7 Jan 2010 21:51:47 +0000 (22:51 +0100)]
[BUG] http: fix for capture memory leak was incorrect
That patch was incorrect because under some circumstances, the
capture memory could be freed by session_free() and then again
by http_end_txn(), causing a double free and an eventual segfault.
The pool use count was also reported wrong due to this bug.
The cleanup code was removed from session_free() to remain only
in http_end_txn().
Willy Tarreau [Thu, 7 Jan 2010 12:35:21 +0000 (13:35 +0100)]
[BUG] http: memory leak with captures when using keep-alive
Hank A. Paulson reported a massive memory leak when using keep-alive
mode. The information he provided made it easy to find that captured
request and response headers were erased but not released when renewing
a request.
Fix the error message by unification and goto, previously we had
two independent lists of supported keywords and were raporting 'stats'
instead of a wrong keyword.
Willy Tarreau [Wed, 6 Jan 2010 23:20:41 +0000 (00:20 +0100)]
[MEDIUM] http: make safer use of the DONT_READ and AUTO_CLOSE flags
Several HTTP analysers used to set those flags to values that
were useful but without considering the possibility that they
were not called again to clean what they did. First, replace
direct flag manipulation with more explicit macros. Second,
enforce a rule stating that any buffer which changes one of
these flags from the default must restore it after completion,
so that other analysers see correct flags.
With both this fix and the previous one about analyser bits,
we should not see any more stuck sessions.
Willy Tarreau [Wed, 6 Jan 2010 23:09:04 +0000 (00:09 +0100)]
[MEDIUM] session: also consider request analysers added during response
A request analyser may very well be added while processing a response
(eg: end of an HTTP keep-alive response). It's very dangerous to only
rely on flags that ought to change in order to loop back, so let's
correctly detect a possible new analyser addition instead of guessing.
Willy Tarreau [Wed, 6 Jan 2010 22:53:24 +0000 (23:53 +0100)]
[MAJOR] session: fix the order by which the analysers are run
With the introduction of keep-alive, we have created situations
where an analyser can add other analysers to the current list,
which are behind it, which have already been processed once, and
which are needed immediately because without them there will be
no more I/O activity. This is typically the case for enabling
reading of a new request after preparing for a new request.
Instead of creating specific cases for some analysers (there was
already one such before), we now use a little bit of algorithmics
to create an ordered bit chain supporting priorities and fast
operations.
Another advantage of this new construction is that it's not a
real loop anymore, so if an analyser is unknown, it will not
loop but just ignore it.
Note that it is easy to skip multiple analysers at once now in
order to speed up the checking a bit. Some test code has shown
a minor gain though.
This change has been carefully re-read and has no direct reason
of causing a regression. However it has been tagged "major"
because the fact that it runs the analysers correctly might
trigger an old sleeping bug somewhere in one of the analysers.
This patch implements default-server support allowing to change
default server options. It can be used in [defaults] or [backend]/[listen]
sections. Currently the following options are supported:
- error-limit
- fall
- inter
- fastinter
- downinter
- maxconn
- maxqueue
- minconn
- on-error
- port
- rise
- slowstart
- weight
[MINOR]: stats: add show-legends to report additional informations
Supported informations, available via "tr/td title":
- cap: capabilities (proxy)
- mode: one of tcp, http or health (proxy)
- id: SNMP ID (proxy, socket, server)
- IP (socket, server)
- cookie (backend, server)
[MINOR] stats: add "a link" & "a href" for sockets
This patch adds add "a link" & "a href" html tags for sockets.
As sockets may have the same name like servers, I decided to
add "+" char (forbidden in names assigned to servers), as a prefix.
Willy Tarreau [Tue, 5 Jan 2010 22:12:12 +0000 (23:12 +0100)]
[BUG] http: check options before the connection header
Commit 0dfdf19b6438c2cea47b1dea0442d65bacfc77cf introduced a
regression because the connection header is now parsed and checked
depending on the configured options, but the options are set after
calling it instead of being set before.
Willy Tarreau [Tue, 5 Jan 2010 10:33:11 +0000 (11:33 +0100)]
[MEDIUM] http: restore the original behaviour of option httpclose
Historically, "option httpclose" has always worked the same way. It
only mangles the "Connection" header in the request and the response
if needed, but does not affect the connection by itself, and ignores
any further data. It is dangerous to change this behaviour without
leaving any other alternative. If an active close is desired, it's
better to make use of "option forceclose" which does exactly what
it intends to do.
So as of now, "option httpclose" will only mangle the headers as
before, and will only affect the connection by itself when combined
with another connection-related option (eg: keepalive or server-close).
Willy Tarreau [Mon, 4 Jan 2010 22:13:26 +0000 (23:13 +0100)]
[BUG] http: some possible missed close remain in the forward chain
We basically have to mimmic the code of process_session() here, so
when the remote output is closed, we must abort otherwise we'll end
up with data which cannot leave the buffer.
Willy Tarreau [Mon, 4 Jan 2010 21:57:43 +0000 (22:57 +0100)]
[BUG] http_process_res_common() must not skip the forward analyser
By default this function returned 0 indicating an end of analysis.
This was not a problem as long as it was the last analyser in the
chain but becomes quite a big one now since it skips the forwarder
with auto_close enabled, causing some data to pass under the nose
of the last one undetected.
Willy Tarreau [Mon, 4 Jan 2010 20:15:02 +0000 (21:15 +0100)]
[MAJOR] http: fix again the forward analysers
There were still several situations leading to CLOSE_WAIT sockets
remaining there forever because some complex transitions were
obviously not caught due to the impossibility to resync changes
between the request and response FSMs.
This patch now centralizes the global transaction state and feeds
it from both request and response transitions. That way, whoever
finishes first, there will be no issue for converging to the correct
state.
Some heavy use of the new debugging function has helped a lot. Maybe
those calls could be removed after some time. First tests are very
positive.
Willy Tarreau [Mon, 4 Jan 2010 20:13:14 +0000 (21:13 +0100)]
[DEBUG] add an http_silent_debug function to debug HTTP states
This function outputs to fd #-1 the status of request and response
buffers, the transaction states, the stream interface states, etc...
That way, it's easy to find that output in an strace report, correctly
placed WRT the other syscalls.
Willy Tarreau [Mon, 4 Jan 2010 06:10:34 +0000 (07:10 +0100)]
[BUG] http: fix hopefully last closing issue on data forwarding
The data forwarders are analysers. As such, the have to check for
various situations on which they have to abort, one of them being
the lack of data with closed input. Now we don't leave the functions
anymore without performing these checks. This has solved the new
CLOSE_WAIT issue that became more noticeable since last patch.
Willy Tarreau [Sun, 3 Jan 2010 22:23:36 +0000 (23:23 +0100)]
[RELEASE] Released version 1.4-dev5
Released version 1.4-dev5 with the following main changes :
- [MINOR] server tracking: don't care about the tracked server's mode
- [MEDIUM] appsession: add "len", "prefix" and "mode" options
- [MEDIUM] appsession: add the "request-learn" option
- [BUG] Configuration parser bug when escaping characters
- [MINOR] CSS & HTML fun
- [MINOR] Collect & provide http response codes received from servers
- [BUG] Fix silly typo: hspr_other -> hrsp_other
- [MINOR] Add "a name" to stats page
- [MINOR] add additional "a href"s to stats page
- [MINOR] Collect & provide http response codes for frontends, fix backends
- [DOC] some small spell fixes and unifications
- [MEDIUM] Decrease server health based on http responses / events, version 3
- [BUG] format '%d' expects type 'int', but argument 5 has type 'long int'
- [BUG] config: fix erroneous check on cookie domain names, again
- [BUG] Healthchecks: get a proper error code if connection cannot be completed immediately
- [DOC] trivial fix for man page
- [MINOR] config: report all supported options for the "bind" keyword
- [MINOR] tcp: add support for the defer_accept bind option
- [MINOR] unix socket: report the socket path in case of bind error
- [CONTRIB] halog: support searching by response time
- [DOC] add a reminder about obsolete documents
- [DOC] point to 1.4 doc, not 1.3
- [DOC] option tcp-smart-connect was missing from index
- [MINOR] http: detect connection: close earlier
- [CLEANUP] sepoll: clean up the fd_clr/fd_set functions
- [OPTIM] move some rarely used fields out of fdtab
- [MEDIUM] fd: merge fd_list into fdtab
- [MAJOR] buffer: flag BF_DONT_READ to disable reads when not required
- [MINOR] http: add new transaction flags for keep-alive and content-length
- [MEDIUM] http request: parse connection, content-length and transfer-encoding
- [MINOR] http request: update the TX_SRV_CONN_KA flag on rewrite
- [MINOR] http request: simplify the test of no-data
- [MEDIUM] http request: simplify POST length detection
- [MEDIUM] http request: make use of pre-parsed transfer-encoding header
- [MAJOR] http: create the analyser which waits for a response
- [MINOR] http: pre-set the persistent flags in the transaction
- [MEDIUM] http response: check body length and set transaction flags
- [MINOR] http response: update the TX_CLI_CONN_KA flag on rewrite
- [MINOR] http: remove the last call to stream_int_return
- [IMPORT] import ebtree v5.0 into directory ebtree/
- [MEDIUM] build: switch ebtree users to use new ebtree version
- [CLEANUP] ebtree: remove old unused files
- [BUG] definitely fix regparm issues between haproxy core and ebtree
- [CLEANUP] ebtree: cast to char * to get rid of gcc warning
- [BUILD] missing #ifndef in ebmbtree.h
- [BUILD] missing #ifndef in ebsttree.h
- [MINOR] tools: add hex2i() function to convert hex char to int
- [MINOR] http: create new MSG_BODY sub-states
- [BUG] stream_sock: BUF_INFINITE_FORWARD broke splice on 64-bit platforms
- [DOC] option is "defer-accept", not "defer_accept"
- [MINOR] http: keep pointer to beginning of data
- [BUG] x-original-to: name was not set in default instance
- [MINOR] http: detect tunnel mode and set it in the session
- [BUG] config: fix error message when config file is not found
- [BUG] config: fix wrong handling of too large argument count
- [BUG] config: disable 'option httplog' on TCP proxies
- [BUG] config: fix erroneous check on cookie domain names
- [BUG] config: cookie domain was ignored in defaults sections
- [MINOR] config: support passing multiple "domain" statements to cookies
- [MINOR] ebtree: add functions to lookup non-null terminated strings
- [MINOR] config: don't report error on all subsequent files on failure
- [BUG] second fix for the printf format warning
- [BUG] check_post: limit analysis to the buffer length
- [MEDIUM] http: process request body in a specific analyser
- [MEDIUM] backend: remove HTTP POST parsing from get_server_ph_post()
- [MAJOR] http: completely process the "connection" header
- [MINOR] http: only consider chunk encoding with HTTP/1.1
- [MAJOR] buffers: automatically compute the maximum buffer length
- [MINOR] http: move the http transaction init/cleanup code to proto_http
- [MINOR] http: move 1xx handling earlier to eliminate a lot of ifs
- [MINOR] http: introduce a new synchronisation state : HTTP_MSG_DONE
- [MEDIUM] http: rework chunk-size parser
- [MEDIUM] http: add a new transaction flags indicating if we know the transfer length
- [MINOR] buffers: add buffer_ignore() to skip some bytes
- [BUG] http: offsets are relative to the buffer, not to ->som
- [MEDIUM] http: automatically re-aling request buffer
- [BUG] http: body parsing must consider the start of message
- [MINOR] new function stream_int_cond_close()
- [MAJOR] http: implement body parser
- [BUG] http: typos on several unlikely() around header insertion
- [BUG] stream_sock: wrong max computation on recv
- [MEDIUM] http: rework the buffer alignment logic
- [BUG] buffers: wrong size calculation for displaced data
- [MINOR] stream_sock: prepare for closing when all pending data are sent
- [MEDIUM] http: add two more states for the closing period
- [MEDIUM] http: properly handle "option forceclose"
- [MINOR] stream_sock: add SI_FL_NOLINGER for faster close
- [MEDIUM] http: make forceclose use SI_FL_NOLINGER
- [MEDIUM] session: set SI_FL_NOLINGER when aborting on write timeouts
- [MEDIUM] http: add some SI_FL_NOLINGER around server errors
- [MINOR] config: option forceclose is valid in frontends too
- [BUILD] halog: insufficient include path in makefile
- [MEDIUM] http: make the analyser not rely on msg being initialized anymore
- [MEDIUM] http: make the parsers able to wait for a buffer flush
- [MAJOR] http: add support for option http-server-close
- [BUG] http: ensure we abort data transfer on write error
- [BUG] last fix was overzealous and disabled server-close
- [BUG] http: fix erroneous trailers size computation
- [MINOR] stream_sock: enable MSG_MORE when forwarding finite amount of data
- [OPTIM] http: set MSG_MORE on response when a pipelined request is pending
- [BUG] http: redirects were broken by chunk changes
- [BUG] http: the request URI pointer is relative to the buffer
- [OPTIM] http: don't immediately enable reading on request
- [MINOR] http: move redirect messages to HTTP/1.1 with a content-length
- [BUG] http: take care of errors, timeouts and aborts during the data phase
- [MINOR] http: don't wait for sending requests to the server
- [MINOR] http: make the conditional redirect support keep-alive
- [BUG] http: fix cookie parser to support spaces and commas in values
- [MINOR] config: some options were missing for "redirect"
- [MINOR] redirect: add support for unconditional rules
- [MINOR] config: centralize proxy struct initialization
- [MEDIUM] config: remove the limitation of 10 reqadd/rspadd statements
- [MEDIUM] config: remove the limitation of 10 config files
- [CLEANUP] http: remove a remaining impossible condition
- [OPTIM] http: optimize a bit the construct of the forward loops
Willy Tarreau [Sun, 3 Jan 2010 18:45:54 +0000 (19:45 +0100)]
[BUG] http: fix cookie parser to support spaces and commas in values
The cookie parser could be fooled by spaces or commas in cookie names
and values, causing the persistence cookie not to be matched if located
just after such a cookie. Now spaces found in values are considered as
part of the value, and spaces, commas and semi-colons found in values
or names, are skipped till next cookie name.
[BUG] Healthchecks: get a proper error code if connection cannot be completed immediately
In case of a non-blocking socket, used for connecting to a remote
server (not localhost), the error reported by the health check
was most of a time one of EINPROGRESS/EAGAIN/EALREADY.
This patch adds a getsockopt(..., SO_ERROR, ...) call so now
the proper error message is reported.
Willy Tarreau [Sun, 3 Jan 2010 16:32:57 +0000 (17:32 +0100)]
[MINOR] http: make the conditional redirect support keep-alive
It makes sense to permit a client to keep its connection when
performing a redirect to the same host. We only detect the fact
that the redirect location begins with a slash to use the keep-alive
(if the client supports it).
Willy Tarreau [Sun, 3 Jan 2010 16:24:51 +0000 (17:24 +0100)]
[MINOR] http: don't wait for sending requests to the server
By default we automatically wait for enough data to fill large
packets if buf->to_forward is not null. This causes a problem
with POST/Expect requests which have a data size but no data
immediately available. Instead of causing noticeable delays on
such requests, simply add a flag to disable waiting when sending
requests.
Willy Tarreau [Sun, 3 Jan 2010 16:07:49 +0000 (17:07 +0100)]
[BUG] http: take care of errors, timeouts and aborts during the data phase
In server-close mode particularly, the response buffer is marked for
no-auto-close after a response passed through. This prevented a POST
request from being aborted on errors, timeouts or anything if the
response was received before the request was complete.
Willy Tarreau [Sun, 3 Jan 2010 13:38:03 +0000 (14:38 +0100)]
[OPTIM] http: don't immediately enable reading on request
If we enable reading of a request immediately after completing
another one, we end up performing small reads until the request
buffer is complete. This takes time and makes it harder to realign
the buffer when needed. Just enable reading when we need to.
Willy Tarreau [Sun, 3 Jan 2010 12:04:35 +0000 (13:04 +0100)]
[BUG] http: the request URI pointer is relative to the buffer
The rq.u field is relative to buf->data, not to msg->sol. We have
to subtract msg->som everywhere this error was made. Maybe it will
be simpler to have a pointer to the buffer in the message and find
appropriate data there.
Willy Tarreau [Sun, 3 Jan 2010 10:37:54 +0000 (11:37 +0100)]
[OPTIM] http: set MSG_MORE on response when a pipelined request is pending
Many times we see a lot of short responses in HTTP (typically 304 on a
reload). It is a waste of network bandwidth to send that many small packets
when we know we can merge them. When we know that another HTTP request is
following a response, we set BF_EXPECT_MORE on the response buffer, which
will turn MSG_MORE on exactly once. That way, multiple short responses can
leave pipelined if their corresponding requests were also pipelined.
Willy Tarreau [Sun, 3 Jan 2010 10:18:34 +0000 (11:18 +0100)]
[MINOR] stream_sock: enable MSG_MORE when forwarding finite amount of data
While it could be dangerous to enable MSG_MORE on infinite data (eg:
interactive sessions), it makes sense to enable it when we know the
chunk to be sent is just a part of a larger one.
We used to forward more trailers than required, causing a
desynchronization of the output. Now we schedule all for forwarding
as soon as we encounter them.
Willy Tarreau [Sat, 2 Jan 2010 21:47:18 +0000 (22:47 +0100)]
[MAJOR] http: add support for option http-server-close
This option enables HTTP keep-alive on the client side and close mode
on the server side. This offers the best latency on the slow client
side, and still saves as many resources as possible on the server side
by actively closing connections. Pipelining is supported on both requests
and responses, though there is currently no reason to get pipelined
responses.
Willy Tarreau [Sat, 2 Jan 2010 21:04:45 +0000 (22:04 +0100)]
[MEDIUM] http: make the parsers able to wait for a buffer flush
When too large a message lies in a buffer before parsing a new
request/response, we can now wait for previous outgoing data to
leave the buffer before attempting to parse again. After that
we can consider the opportunity to realign the buffer if needed.
Willy Tarreau [Sat, 2 Jan 2010 20:59:16 +0000 (21:59 +0100)]
[MEDIUM] http: make the analyser not rely on msg being initialized anymore
The HTTP parser needed the msg structure to hold pre-initialized pointers.
This causes a trouble with keep-alive because if some data is still in the
buffer, the pointers can be anywhere after the data and later become invalid
when the buffer gets realigned.
It was not needed to rely on that since we have two valid information
in the buffer itself :
- buf->lr : last visited place
- buf->w + buf->send_max : beginning of next message
So by doing the maths only on those values, we can avoid doing tricks
on msg->som.
Willy Tarreau [Wed, 30 Dec 2009 00:10:35 +0000 (01:10 +0100)]
[MINOR] config: option forceclose is valid in frontends too
This option was disabled for frontends in the configuration because
it was useless in its initial implementation, though it was still
checked in the code. Let's officially enable it now.
Willy Tarreau [Tue, 29 Dec 2009 13:56:36 +0000 (14:56 +0100)]
[MEDIUM] http: add some SI_FL_NOLINGER around server errors
When we catch an error from the server, speed up the connection
abort since we don't want to remain long with pending data in the
socket, and we want to be able to reuse our source port ASAP.
Willy Tarreau [Tue, 29 Dec 2009 13:49:56 +0000 (14:49 +0100)]
[MEDIUM] session: set SI_FL_NOLINGER when aborting on write timeouts
Doing this helps us flush the system buffers from all unread data. This
avoids having orphans when clients suddenly get off the net without
reading their entire response.
Willy Tarreau [Tue, 29 Dec 2009 13:36:34 +0000 (14:36 +0100)]
[MINOR] stream_sock: add SI_FL_NOLINGER for faster close
This new flag may be set by any user on a stream interface to tell
the underlying protocol that there is no need for lingering on the
socket since we know the other side either received everything or
does not care about what we sent.
This will typically be used with forced server close in HTTP mode,
where we want to quickly close a server connection after receiving
its response. Otherwise the system would prevent us from reusing
the same port for some time.
The "forceclose" option used to close the output channel to the
server once it started to respond. While this happened to work with
most servers, some of them considered this as a connection abort and
immediately stopped responding.
Now that we're aware of the end of a request and response, we're able
to trivially handle this option and properly close both sides when the
server's response is complete.
During this change it appeared that forwarding could be allowed when
the BF_SHUTW_NOW flag was set on a buffer, which obviously is not
acceptable and was causing some trouble. This has been fixed too and
is the reason for the MEDIUM status on this patch.
Willy Tarreau [Tue, 29 Dec 2009 07:02:56 +0000 (08:02 +0100)]
[MINOR] stream_sock: prepare for closing when all pending data are sent
Since we'll soon be able to close a connection with remaining data in a
buffer, it becomes obvious that we can prepare to close when we're about
to send the last chunk of data and not the whole buffer.
Willy Tarreau [Mon, 28 Dec 2009 17:37:54 +0000 (18:37 +0100)]
[BUG] buffers: wrong size calculation for displaced data
This error was triggered by requests not starting at the beginning
of the buffer. It cannot happen with earlier versions though it might
be a good idea to fix it anyway.
Willy Tarreau [Mon, 28 Dec 2009 16:39:57 +0000 (17:39 +0100)]
[MEDIUM] http: rework the buffer alignment logic
There were still issues with the buffer alignment. Now we ensure
that we always align it before a request or response is completely
parsed if there is less than maxrewrite bytes free at the end. In
practice, it's not called that often and ensures we can always work
as expected.
Willy Tarreau [Mon, 28 Dec 2009 16:36:37 +0000 (17:36 +0100)]
[BUG] stream_sock: wrong max computation on recv
Since the introduction of the automatic sizing of buffers during reads,
a bug appeared where the max size could be negative, causing large
chunks of memory to be overwritten during recv() calls if a read pointer
was already past the buffer's limit.
Willy Tarreau [Sun, 27 Dec 2009 21:54:55 +0000 (22:54 +0100)]
[MAJOR] http: implement body parser
The body parser will be used in close and keep-alive modes. It follows
the stream to keep in sync with both the request and the response message.
Both chunked transfer-coding and content-length are supported according to
RFC2616.
The multipart/byterange encoding has not yet been implemented and if not
seconded by any of the two other ones, will be forwarded till the close,
as requested by the specification.
Both the request and the response analysers converge into an HTTP_MSG_DONE
state where it will be possible to force a close (option forceclose) or to
restart with a fresh new transaction and maintain keep-alive.
This change is important. All tests are OK but any possible behaviour
change with "option httpclose" might find its root here.
Willy Tarreau [Sun, 27 Dec 2009 21:47:25 +0000 (22:47 +0100)]
[BUG] http: body parsing must consider the start of message
When parsing body for URL parameters, we must not consider that
data are available from buf->data but from buf->data + msg->som.
This is not a problem right now but may become with keep-alive.
When parsing a request that does not start at the beginning of the
buffer, we may experience a buffer full issue. In order to avoid
this, we try to realign the buffer if it is not really full. That
will be required when we have to deal with pipelined requests.
Willy Tarreau [Sun, 27 Dec 2009 14:50:06 +0000 (15:50 +0100)]
[BUG] http: offsets are relative to the buffer, not to ->som
Some wrong operations were performed on buffers, assuming the
offsets were relative to the beginning of the request while they
are relative to the beginning of the buffer. In practice this is
not yet an issue since both are the same... until we add support
for keep-alive.
Willy Tarreau [Sat, 26 Dec 2009 14:34:26 +0000 (15:34 +0100)]
[MEDIUM] http: add a new transaction flags indicating if we know the transfer length
It's not enough to know if the connection will be in CLOSE or TUNNEL mode,
we still need to know whether we want to read a full message to a known
length or read it till the end just as in TUNNEL mode. Some updates to the
RFC clarify slightly better the corner cases, in particular for the case
where a non-chunked encoding is used last.
Now we also take care of adding a proper "connection: close" to messages
whose size could not be determined.