smp_fetch_srv_name() stored a raw pointer to srv->id in the sample
without setting SMP_F_CONST. Every other sibling id-pointer fetch
(smp_fetch_be_name on px->id, smp_fetch_fe_name on fe->id, the SSL
helpers using OBJ_nid2sn() / SSL_get_cipher_name(), etc.) correctly
sets SMP_F_CONST to prevent in-place mutation by converters such as
,upper / ,lower / ,regsub.
Without SMP_F_CONST, an expression like srv_name,lower would write
into srv->id for the lifetime of the process. In practice this has
gone unnoticed because srv->id is a private allocation that is never
read back by name, but the bug is real and the divergence from the
other id fetches is unintentional.
This becomes more important with the introduction of runtime server
renaming (next patch in series): SMP_F_CONST ensures that callers go
through smp_make_rw() / smp_dup() before mutating, isolating the
sample's bytes from the server's id storage.
This is a stand-alone fix and should be backported.
These ones were deprecated in 3.3-dev2 with commits 5c15ba5eff ("MEDIUM:
proxy: mark the "dispatch" directive as deprecated") and e93f3ea3f8
("MEDIUM: proxy: deprecate the "transparent" and "option transparent"
directives"), and were planned for removal in 3.5. See also:
https://github.com/orgs/haproxy/discussions/2921
as well as the wiki page about breaking changes.
They've lived their lives and always cause internal limitations
(exceptions between connecting to server or connecting to proxy), and
are even confusing to some extents (especially "transparent" which users
often get wrong).
This commit removes the ability to configure them, tests based on them
and all the doc related to them. The keywords remain detected by the
parser and indicate how to proceed instead.
It's likely that other deeper parts will be changed as well (e.g.
conn->target will no longer be of OBJ_TYPE_PROXY). This will be done
over the long term.
Use xprt_add_l6hs() at the end of connect_server() if selected MUX layer
relies on a temporary handshake prior to its initialization. This
functions is noop is SSL layer is active.
This change is necessary to support clear QMux on the backend side.
Recently defined <init_xprt> from mux_proto_list is used to render the
code as generic as possible.
When we store the negociated server parameters, such as the ALPN, also
store the calculated hash with the connection. If it is different, as
can happen because the IP address is different because set-dst was used,
we certainly do not want to reuse the information in the cache,
otherwise we could end up using the wrong ALPN and mux.
That means we already have to calculate the hash in connect_server()
now, while before we would not do it for Websockets, if we could not do
connection reuse, as that's all the hash was used for.
This should fix Github issue #3386
This should be backported as far as 3.2.
The "hash-type xxx none" is broken for keys that are not in type string
because the sample fetch call casts them to SMP_T_BIN, that tends to
preserve the original format (integers, IP addresses etc), but the
gen_hash() function in case of BE_LB_HFCN_NONE expects to read a string
representing a number, that it parses to retrieve the value, and just
fails on many binary types. For example, the following just always
returns key 0:
balance hash rand()
hash type consistent none
An ugly workaround is to make sure the expression returns a string, for
example this:
balance hash rand(),concat()
hash type consistent none
In order to fix most cases here, we force the conversion to type string
when using BE_LB_HFCN_NONE, but a better approach would require a larger
rework and split gen_hash() or change it to accept an integer as well,
so that the caller could cast to SMP_T_INT for BE_LB_HFCN_NONE and pass
the resulting number already parsed with the least information loss. In
this case even IPv4 addresses would be preserved.
The current approach at least addresses the initially envisioned use
cases, and the limitations have been added to the doc. This can be
backported to 3.0 though it's not really important.
The first part of this patch defines a new mux_proto_list field named
<xprt_init>. This allows to define an extra XPRT layer which should be
activated first prior to the MUX creation both on frontend and backend
sides.
This is immediately used for QMux mux_proto_list to require XPRT_QMUX
handshake. With this change, activation of QMux connection flags in
session_accept_fd() and connect_server() are adjusted to take into
account <init_xprt> field. This approach is much more evolutive than
relying on the previous MUX name.
Change in connect_server() will also be necessary to support QMux
activation on a TCP server with h3 ALPN without explicit "proto qmux".
This guarantees that MUX initialization is delayed after QMux handshake.
In struct mux_proto_list, rename the "token" field to "mux_proto". That
field should only be used to match the name provided in the "proto"
directive, and it will be soon.
This should be a no-op.
In the inner while loop that validates each character of a POST parameter
value, the code checks *p via HTTP_IS_TOKEN() and HTTP_IS_LWS() instead
of *end, while the loop condition only advances "end", so only the first
character of each value is validated.
This means spaces or binary data embedded in parameter values after the
first character goes undetected. Fix by replacing both references to *p
with *end to properly scan through all characters as intended.
This bug was introduced in 1.5-dev20 by commit 98634f0c7 ("MEDIUM:
backend: Enhance hash-type directive with an algorithm options") so
the fix must be backported to all versions.
This is a follow-up on the QUIC MUX renaming process.
The current patch performs renaming of "qstrm" to "qmux" in connection
flags. These flags are only used in linked with the xprt_qmux layer.
This has an impact on every files which manipulates these flags, namely
backend, session and ssl_sock sources.
Also, internal xprt identifier is renamed from XPRT_QSTRM to XPRT_QMUX,
This lets lb_ops specify the conditions necessary to bind to this set of
ops. The condition is expressed as a list of mask and match fields on
the algorithm flags. This is then used in proxy_finalize() to locate the
lb_ops corresponding to the current configuration, by iterating over
the list of lb_ops structures. This list is implemented using the same
mechanisms used for configuration keywords: an INITCALL1 macro to a
registration function.
This also moves the lookup and property flags into the lb_ops structure
that were previously applied manually on a case by case basis.
The HTTP transaction is moved in an union. For now, it is the only possible
transaction that can be allocated. But that will change. Thanks to this
commit and the next one, it will be possible to deal with different kind of
transactions for a stream.
This patch looks quite huge, but it is more or less a renaming of all
accesses to "txn" field by "txn.http".
When an idle connection is private or considered private, session_add_conn()
is called to add it to the list of connections owned by the session. But
in case of allocation failure, the session is not set, which results in
a long list of possible situations that are all corner cases which are
difficult to test (and debug).
This commit relies on the fact that it is already permitted to have
conn->owner pointing to a session even if the connection couldn't be
added to the session's list, as this was already the case in
conn_backend_get() when dealing with HOL_RISK. Also as seen in commit
3aab17bd56 added in 2.4, it is already possible to have conn->owner
set with the connection not being in a list, and only the list element
is checked for this.
This commit modifies session_add_conn() to always set conn->onwer, even
if the list element couldn't be allocated. This way it's possible to
always refer to conn->owner to find the session owning a private conn
even in case of failure to allocate an entry. This requires to change
the checks on conn->owner to a check of the list element to see if the
connection belongs to a session, the pre-assignment of sess to
conn->owner in conn_backend_get() is no longer needed, same for the
pre-assignment in http_wait_for_response(), and that's all.
The H1 mux remained unchanged because since it cannot multiplex, in
case it fails to allocate a pconn, it instantly kills the connection.
Most calls to mux ops were instrumented with a CALL_MUX_WITH_RET() or
CALL_MUX_NO_RET() macro in order to make the current thread's context
point to the called mux and be able to track its allocations. Only
a bunch of harmless mux_ctl() and ->subscribe/unsubscribe calls were
left untouched since useless. But destroy/detach/shut/init/snd_buf
and rcv_buf are now tracked.
It will not show allocations performed in IO callback via tasklet
wakeups however.
In order to ease reading of the output, cmp_memprof_ctx() knows about
muxes and sorts based on the .subscribe function address instead of
the mux_ops address so as to keep various callers grouped.
In connect_server(), it is possible to have no server defined (dispatch mode
or transparent backend). In that case, we must be carefull to check the srv
variable in all calls involving the server. It was not perform at one place,
when the protocol to use for websocket is retrieved. This must not be done
when there is no server.
This patch should fix the first report in #3144. It must be backported to
all stable version.
Introduce COUNTERS_UPDATE_MAX(), and use it instead of using
HA_ATOMIC_UPDATE_MAX() directly.
For now it just calls HA_ATOMIC_UPDATE_MAX(), but will later be modified
so that we can disable max calculation.
This can be backported up to 2.8 if the usage of COUNTERS_UPDATE_MAX()
generates too many conflicts.
In connect_server(), MUX initialization must be delayed if ALPN
negotiation is configured, unless ALPN can already be retrieved via the
server cache.
A readlock is used to consult the server cache. Prior to this patch, it
was always taken even if no ALPN is configured. The lock was thus used
for every new backend connection instantiation.
Rewrite the check so that now the lock is only used if ALPN is
configured. Thus, no lock access is done if SSL is not used or if ALPN
is not defined.
In practice, there will be no performance gain, as the read lock should
never block if ALPN is not configured. However, the code is cleaner as
it better reflect that only access to server nego_alpn requires the
path_params lock protection.
In connect_server(), when a new connection must be instantiated, MUX
initialization is delayed if an ALPN setting is present on the server
line configuration, as negotiation must be performed to select the
correct MUX. However, this is not the case if the ALPN can already be
retrieved on the server cache.
This check is performed too late however and may cause issue with the
QUIC stack. The problem can happen when the server ALPN is not yet set.
In the normal case, quic_conn layer is instantiated and MUX init is
delayed until the handshake completion. When the MUX is finally
instantiated, it reused without any issue app_ops from its quic_conn,
which is derived from the negotiated ALPN.
However, there is a race condition if another QUIC connection populates
the server ALPN cache. If this happens after the first quic_conn init
but prior to the MUX delay check, the MUX will thus immediately start in
connect_server(). When app_ops is retrieved from its quic_conn, a crash
occurs in qcc_install_app_ops() as the QUIC handshake is not yet
finalized :
#0 0x000055e242a66df4 in qcc_install_app_ops (qcc=0x7f127c39da90, app_ops=0x0) at src/mux_quic.c:1697
1697 if (app_ops->init && !app_ops->init(qcc)) {
[Current thread is 1 (Thread 0x7f12810f06c0 (LWP 25758))]
To fix this, MUX delay check is moved up in connect_server(). It is now
performed prior conn_prepare() which is responsible for the quic_conn
layer instantiation. Thus, it ensures consistency for the QUIC stack :
MUX init is always delayed if the quic_conn does not reuses itself the
SSL session and ALPN server cache (no quic_reuse_srv_params()).
This must be backported up to 3.3.
Add the ability to set connect, queue and tarpit timeouts from the
set-timeout action. This is especially useful when using set-dst to
dynamically connect to servers.
This patch also adds the relevant fe_/be_/cur_ sample fetches for these
timeouts.
This is a follow up to b6bdb2553 ("MEDIUM: backend: make "balance random"
consider req rate when loads are equal")
In the above patch, we used the global sess_per_sec metric to choose which
server we should be using. But the original intent was to use the per
thread group statistic.
No backport needed, the previous patch already improved the situation in
3.3, so let's not take the risk of breaking that.
Move backend compatibility checks performed during 'add server' in a
dedicated function be_supports_dynamic_srv(). This should simplify
addition of future restriction.
This function will be reused when implementing backend creation at
runtime.
As reported by Damien Claisse and Cdric Paillet, the "random" LB
algorithm can become particularly unfair with large numbers of servers
having few connections. It's indeed fairly common to see many servers
with zero connection in a thousand-server large farm, and in this case
the P2C algo consisting in checking the servers' loads doesn't help at
all and is basically similar to random(1). In this case, we only rely
on the distribution of server IDs in the random space to pick the best
server, but it's possible to observe huge discrepancies.
An attempt to model the problem clearly shows that with 1600 servers
with weight 10, for 1 million requests, the lowest loaded ones will
take 300 req while the most loaded ones will get 780, with most of
the values between 520 and 700.
In addition, only the first 28 lower bits of server IDs are used for
the key calculation, which means that node keys are more determinist.
Setting random keys in the lowest 28 bits only better packs values
with min around 530 and max around 710, with values mostly between
550 and 680.
This can only be compensated by increasing weights and draws without
being a perfect fix either. At 4 draws, the min is around 560 and the
max around 670, with most values bteween 590 and 650.
This patch takes another approach to this problem: when servers are on
tie regarding their loads, instead of arbitrarily taking the second one,
we now compare their current request rates, which is updated all the
time and smoothed over one second, and we pick the server with the
lowest request rate. Now with 2 draws, the curve is mostly flat, with
the min at 580 and the max at 628, and almost all values between 611
and 625. And 4 draws exclusively gives values from 614 to 624.
Other points will need to be addressed separately (bits of server ID,
maybe refine the hash algorithm), but these ones would affect how
caches are selected, and cannot be changed without an extra option.
For random however we can perform a change without impacting anyone.
This should be backported, probably only to 3.3 since it's where the
"random" algo became the default.
Contrarily to what was previously believed, there are corner cases where
the counters may not be allocated, and we may want to make them optional
at a later date, so we have to check if those counters are there.
However, just checking that shared.tg is non-NULL is enough, we can then
assume that shared.tg[tgid - 1] has properly been allocated too.
Also modify the various COUNTER_SHARED_* macros to make sure they check
for that too.
Before updating counters, a few tests are made to check if the counters
exits. but those counters should always exist at this point, so just
remmove them.
This commit should have no impact, but can easily be reverted with no
functional impact if various crashes appear.
Instead of statically allocating the per-thread group counters,
based on the max number of thread groups available, allocate
them dynamically, based on the number of thread groups actually
used. That way we can increase the maximum number of thread
groups without using an unreasonable amount of memory.
In 2.6, do_connect_server() was introduced by commit 0a4dcb65f ("MINOR:
stream-int/backend: Move si_connect() in the backend scope") and changed
the approach to work with a stream instead of a stream-interface. However
si_oc(si) was wrongly turned to &s->res instead of &s->req, which breaks
TFO by always inspecting the response channel to figure whether there are
data pending.
This fix can be backported to all versions till 2.6.
In 2.6, the retries counter on a stream was changed from retries left
to retries done via commit 731c8e6cf ("MINOR: stream: Simplify retries
counter calculation"). However, one comparison fell through the cracks
in order to detect whether or not we can use TFO (only first attempt),
resulting in TFO never working anymore.
This may be backported to all versions till 2.6.
Back in the mists of time, commit e91a526c8f decided that if we were trying
to stay on the same server than the previous request, and if there were
a connection available in the session, we'd remove its CO_FL_SESS_IDLE.
The reason for doing that has been long lost, probably it fixed a bug at some
point, but it was most probably not the right place to do that. And starting
with 3.3, this triggers a BUG_ON() because that flag is expected later on.
So just revert the commit, if the ancient bug shows up again, it will be
fixed another way.
This should be backported to 3.3. There is little reason to backport it
to previous versions, unless other patches depend on it.
The server's xprt is always defined and cannot be NULL. So there is no
reason to test it. It could lead to wrong assumptions later in the code.
This patch should fix a Coverity report from #3213.
The SNI of a new connection is now retrieved earlier, before the
initialization of the SSL context. So, concretely, it is now performed
before calling conn_prepare(). The SNI is then set just after.
When a SNI is set on a new connection, its hash is now saved in the
connection itself. To do so, a dedicated field was added into the connection
strucutre, called sni_hash. For now, this value is only used when the TLS
session is cached.
This reverts commit de29000e60.
The fix was in fact invalid. First it is not supprted by WolfSSL to call
SSL_set_tlsext_host_name with a hostname to NULL. Then, it is not specified
as supported by other SSL libraries.
But, by reviewing the root cause of this bug, it appears there is an issue
with the reuse of TLS sesisons. It must not be performed if the SNI does not
match. A TLS session created with a SNI must not be reused with another
SNI. The side effects are not clear but functionnaly speaking, it is
invalid.
So, for now, the commit above was reverted because it is invalid and it
crashes with WolfSSL. Then the init of the SSL connection must be reworked
to get the SNI earlier, to be able to reuse or not an existing TLS
session.
When a new SSL server connection is created, if no SNI is set, it is
possible to inherit from the one of the reused TLS session. The bug was
introduced by the commit 95ac5fe4a ("MEDIUM: ssl_sock: always use the SSL's
server name, not the one from the tid"). The mixup is possible between
regular connections but also with health-checks connections.
To fix the issue, when no SNI is set, for regular server connections and for
health-check connections, the SNI must explicitly be disabled by calling
ssl_sock_set_servername() with the hostname set to NULL.
Many thanks to Lukas for his detailed bug report.
This patch should fix the issue #3195. It must be backported as far as 3.0.
In assign_server_and_queue(), there's a rare case when the server was
full, so we created a pendconn, another server was considered but in the
meanwhile the pendconn was unqueued already, so we just left the
function. We did so, however, while still holding the queue lock, which
will ultimately lead to a deadlock, and ultimately the watchdog would
kill the process.
To fix that, just unlock the queue before leaving.
This should be backported to 3.2.
The target patch fixes a rare race condition which happen when a MUX IO
handler is working on a connection already moved into the purge list. In
this case, the handler will incorrectly moved back the connection into
the idle list.
To fix this, conn_delete_from_tree() was extended to remove flags along
with the connection from the idle list. This was performed when the
connection is moved into the purge list. However, it introduces another
issue related to the idle server connection accounting. Thus it is
necessary to revert it prior to the incoming newer fix.
This patch must be backported to every version where the original commit
is.
check-reuse-pool can only perform as expected if reuse policy on the
backend is set to aggressive or higher. Update the documentation to
reflect this and implement a server diag warning.
In connect_server(), defer the call to conn_xprt_start() until after we
had a chance to create the mux. The xprt can behave differently
depending on if a mux is or is not available at this point, as if it is,
it may want to wait until some data comes from the mux.
This does not need to be backported.
There's currently a function conn_delete_from_tree() which is used to
detach an idle connection from the tree it's currently attached to so
that it is no longer found. This function is used in three circumstances:
- when picking a new connection that no longer has any avail stream
- when temporarily working on the connection from an I/O handler,
in which case it's re-added at the end
- when killing a connection
The 2nd case above is quite specific, as it requires to preserve the
CO_FL_LIST_MASK flags so that the connection can be re-inserted into
the proper tree when leaving the handler. However, there's a catch.
When killing a connection, we want to be certain it will not be
reinserted into the tree. The flags preservation is causing a tiny
race if an I/O happens while the connection is in the kill list,
because in this case the I/O handler will note the connection flags,
do its work, then reinsert the connection where it believed it was,
then the connection gets purged, and another user can find it in the
tree.
The issue is very difficult to reproduce. On a 128-thread machine it
happens in H2 around 500k req/s after around 50M requests. In H1 it
happens after around 1 billion requests.
The fix here consists in passing an extra argument to the function to
indicate if the removal is permanent or not. When it's permanent, the
function will clear the associated flags. The callers were adjusted
so that all those dequeuing a connection in order to kill it do it
permanently and all other ones do it only temporarily.
A slightly different approach could have worked: the function could
always remove all flags, and the callers would need to restore them.
But this would require trickier modifications of the various call
places, compared to only passing 0/1 to indicate the permanent status.
This will need to be backported to all stable versions. The issue was
at least reproduced since 3.1 (not tested before). The patch will need
to be adjusted for 3.2 and older, because a 2nd argument "thr" was
added in 3.3, so the patch will not apply to older versions as-is.
Add a rwlock to control the server's path_parameter, to make sure
multiple threads don't set it at the same time, and it can't be seen in
an inconsistent state.
Also don't set the parameter every time, only set them if they have
changed, to prevent needless writes.
This does not need to be backported.
When trying to reuse a backend connection, a connection hash is
calculated to match an entry with similar parameters. Previously, this
operation was skipped if the stream content wasn't based on HTTP, as it
would have been incompatible with http-reuse.
With the introduction of SPOP backends, this condition was removed, so
that it can also benefit from connection reuse. However, this means that
now hash calcul is always performed when connecting to a server, even
for TCP or log backends. This is unnecessary as these proxies cannot
perform connection reuse.
Note also that reuse mode is resetted on postparsing for incompatible
backends. This at least guarantees that no tree lookup will be performed
via be_reuse_connection(). However, connection lookup is still performed
in the session via session_get_conn() which is another unnecessary
operation.
Thus, this patch restores the condition so that reuse operations are now
entirely skipped if a backend mode is incompatible. This is implemented
via a new utility function named be_supports_conn_reuse().
This could be backported up to 3.1, as this commit could be considered
as a performance regression for tcp/log backend modes.
In order to prepare for changing the way abortonclose works, let's
replace the direct flag check with a similarly named function
(proxy_abrt_close) which returns the on/off status of the directive
for the proxy. For now it simply reflects the flag's state.
In connect_server(), only avoid creating a mux when we're reusing a
connection, if that connection already has one. We can reuse a
connection with no mux, if we made a first attempt at connecting to the
server and it failed before we could create the mux (or during the mux
creation). The connection will then be reused when trying again.
This fixes a bug where a stream could stall if the first connection
attempt failed before the mux creation. It is easy to reproduce by
creating random memory allocation failure with -dmFail.
This was introduced by commit 4aaf0bfbce,
and thus does not need any backport as long as that commit is not
backported.
There is currently an srv_queue converter which is capable of taking the
output of a dynamic name and determining the queue length for a given
server. In addition there is a sample fetcher for whether a server is
currently up. This simply combines the two such that srv_is_up can be
used as a converter too.
Future work might extend this to other sample fetchers for servers, but
this is probably the most useful for acl routing.
In preparation of providing further server converters, split the code
for finding the server from the sample out.
Additionally, update the documentation for srv_queue converter to note
security concerns.
This patch looks huge, but it has a very simple goal: protect all
accessed to shared stats pointers (either read or writes), because
we know consider that these pointers may be NULL.
The reason behind this is despite all precautions taken to ensure the
pointers shouldn't be NULL when not expected, there are still corner
cases (ie: frontends stats used on a backend which no FE cap and vice
versa) where we could try to access a memory area which is not
allocated. Willy stumbled on such cases while playing with the rings
servers upon connection error, which eventually led to process crashes
(since 3.3 when shared stats were implemented)
Also, we may decide later that shared stats are optional and should
be disabled on the proxy to save memory and CPU, and this patch is
a step further towards that goal.
So in essence, this patch ensures shared stats pointers are always
initialized (including NULL), and adds necessary guards before shared
stats pointers are de-referenced. Since we already had some checks
for backends and listeners stats, and the pointer address retrieval
should stay in cpu cache, let's hope that this patch doesn't impact
stats performance much.
Previously the conn_hash_node was placed outside the connection due
to the big size of the eb64_node that could have negatively impacted
frontend connections. But having it outside also means that one
extra allocation is needed for each backend connection, and that one
memory indirection is needed for each lookup.
With the compact trees, the tree node is smaller (16 bytes vs 40) so
the overhead is much lower. By integrating it into the connection,
We're also eliminating one pointer from the connection to the hash
node and one pointer from the hash node to the connection (in addition
to the extra object bookkeeping). This results in saving at least 24
bytes per total backend connection, and only inflates connections by
16 bytes (from 240 to 256), which is a reasonable compromise.
Tests on a 64-core EPYC show a 2.4% increase in the request rate
(from 2.08 to 2.13 Mrps).
Idle connection trees currently require a 56-byte conn_hash_node per
connection, which can be reduced to 32 bytes by moving to ceb64. While
ceb64 is theoretically slower, in practice here we're essentially
dealing with trees that almost always contain a single key and many
duplicates. In this case, ceb64 insert and lookup functions become
faster than eb64 ones because all duplicates are a list accessed in
O(1) while it's a subtree for eb64. In tests it is impossible to tell
the difference between the two, so it's worth reducing the memory
usage.
This commit brings the following memory savings to conn_hash_node
(one per backend connection), and to srv_per_thread (one per thread
and per server):
struct before after delta
conn_hash_nodea 56 32 -24
srv_per_thread 96 72 -24
The delicate part is conn_delete_from_tree(), because we need to
know the tree root the connection is attached to. But thanks to
recent cleanups, it's now clear enough (i.e. idle/safe/avail vs
session are easy to distinguish).
