[PATCH net-next v2] netfilter: conntrack: expose gc_scan_interval_max via sysctl

[PATCH net-next v2] netfilter: conntrack: expose gc_scan_interval_max via sysctl

[Prasanna S Panchamukhi]

The conntrack garbage collection worker uses an adaptive algorithm that
adjusts the scan interval based on the average timeout of tracked
entries.  The upper bound of this interval is hardcoded as
GC_SCAN_INTERVAL_MAX (60 seconds).

Expose the upper bound as a new sysctl,
net.netfilter.nf_conntrack_gc_scan_interval_max, so it can be tuned at
runtime without rebuilding the kernel.  The default remains 60 seconds
to preserve existing behavior.  The sysctl is global and read-only in
non-init network namespaces, consistent with nf_conntrack_max and
nf_conntrack_buckets.

In environments where long-lived offloaded flows dominate the table,
the adaptive average drifts toward the maximum, delaying cleanup
of short-lived expired entries such as those in TCP CLOSE state
(10s timeout). Adding sysctl to set the maximum GC scan helps to
tune according to the evironment.

Signed-off-by: Prasanna S Panchamukhi <panchamukhi@arista.com>
cc: "David S. Miller" <davem@davemloft.net>
cc: Eric Dumazet <edumazet@google.com>
cc: Jakub Kicinski <kuba@kernel.org>
cc: Paolo Abeni <pabeni@redhat.com>
cc: Simon Horman <horms@kernel.org>
cc: Jonathan Corbet <corbet@lwn.net>
cc: Shuah Khan <skhan@linuxfoundation.org>
cc: Pablo Neira Ayuso <pablo@netfilter.org>
cc: Florian Westphal <fw@strlen.de>
cc: Phil Sutter <phil@nwl.cc>
cc: netdev@vger.kernel.org
cc: linux-doc@vger.kernel.org
cc: linux-kernel@vger.kernel.org
to: netfilter-devel@vger.kernel.org
cc: coreteam@netfilter.org
---
 Documentation/networking/nf_conntrack-sysctl.rst | 13 +++++++++++++
 include/net/netfilter/nf_conntrack.h             |  1 +
 net/netfilter/nf_conntrack_core.c                | 10 +++++++---
 net/netfilter/nf_conntrack_standalone.c          | 10 ++++++++++
 4 files changed, 31 insertions(+), 3 deletions(-)

diff --git a/Documentation/networking/nf_conntrack-sysctl.rst b/Documentation/networking/nf_conntrack-sysctl.rst
index 35f889259fcd..0e79f6ad1062 100644
--- a/Documentation/networking/nf_conntrack-sysctl.rst
+++ b/Documentation/networking/nf_conntrack-sysctl.rst
@@ -64,6 +64,19 @@ nf_conntrack_frag6_timeout - INTEGER (seconds)
 
 	Time to keep an IPv6 fragment in memory.
 
+nf_conntrack_gc_scan_interval_max - INTEGER (seconds)
+	default 60
+
+	Maximum interval between garbage collection scans of the connection
+	tracking table. The GC worker uses an adaptive algorithm that adjusts
+	the scan interval based on average entry timeouts; this parameter caps
+	the upper bound. Lower values cause expired entries (e.g. connections
+	in CLOSE state) to be cleaned up faster, at the cost of slightly more
+	CPU usage. Consider tuning this on systems with high connection churn
+	(e.g. NAT gateways, load balancers) where expired entries accumulate
+	and cause the conntrack table to fill up. Minimum value is 1.
+	This sysctl is only writeable in the initial net namespace.
+
 nf_conntrack_generic_timeout - INTEGER (seconds)
 	default 600
 
diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index bc42dd0e10e6..0449577f322e 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -331,6 +331,7 @@ extern struct hlist_nulls_head *nf_conntrack_hash;
 extern unsigned int nf_conntrack_htable_size;
 extern seqcount_spinlock_t nf_conntrack_generation;
 extern unsigned int nf_conntrack_max;
+extern unsigned int nf_conntrack_gc_scan_interval_max;
 
 /* must be called with rcu read lock held */
 static inline void
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 27ce5fda8993..8647e6824cec 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -91,7 +91,7 @@ static DEFINE_MUTEX(nf_conntrack_mutex);
  * allowing non-idle machines to wakeup more often when needed.
  */
 #define GC_SCAN_INITIAL_COUNT	100
-#define GC_SCAN_INTERVAL_INIT	GC_SCAN_INTERVAL_MAX
+#define GC_SCAN_INTERVAL_INIT	READ_ONCE(nf_conntrack_gc_scan_interval_max)
 
 #define GC_SCAN_MAX_DURATION	msecs_to_jiffies(10)
 #define GC_SCAN_EXPIRED_MAX	(64000u / HZ)
@@ -204,6 +204,9 @@ EXPORT_SYMBOL_GPL(nf_conntrack_htable_size);
 
 unsigned int nf_conntrack_max __read_mostly;
 EXPORT_SYMBOL_GPL(nf_conntrack_max);
+
+unsigned int nf_conntrack_gc_scan_interval_max __read_mostly = GC_SCAN_INTERVAL_MAX;
+
 seqcount_spinlock_t nf_conntrack_generation __read_mostly;
 static siphash_aligned_key_t nf_conntrack_hash_rnd;
 
@@ -1515,6 +1518,7 @@ static void gc_worker(struct work_struct *work)
 	unsigned int i, hashsz, nf_conntrack_max95 = 0;
 	u32 end_time, start_time = nfct_time_stamp;
 	struct conntrack_gc_work *gc_work;
+	unsigned long gc_scan_max = READ_ONCE(nf_conntrack_gc_scan_interval_max);
 	unsigned int expired_count = 0;
 	unsigned long next_run;
 	s32 delta_time;
@@ -1568,7 +1572,7 @@ static void gc_worker(struct work_struct *work)
 				delta_time = nfct_time_stamp - gc_work->start_time;
 
 				/* re-sched immediately if total cycle time is exceeded */
-				next_run = delta_time < (s32)GC_SCAN_INTERVAL_MAX;
+				next_run = delta_time < (s32)gc_scan_max;
 				goto early_exit;
 			}
 
@@ -1630,7 +1634,7 @@ static void gc_worker(struct work_struct *work)
 
 	gc_work->next_bucket = 0;
 
-	next_run = clamp(next_run, GC_SCAN_INTERVAL_MIN, GC_SCAN_INTERVAL_MAX);
+	next_run = clamp(next_run, GC_SCAN_INTERVAL_MIN, gc_scan_max);
 
 	delta_time = max_t(s32, nfct_time_stamp - gc_work->start_time, 1);
 	if (next_run > (unsigned long)delta_time)
diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c
index 207b240b14e5..f8cab779763f 100644
--- a/net/netfilter/nf_conntrack_standalone.c
+++ b/net/netfilter/nf_conntrack_standalone.c
@@ -637,6 +637,7 @@ enum nf_ct_sysctl_index {
 	NF_SYSCTL_CT_PROTO_TIMEOUT_GRE,
 	NF_SYSCTL_CT_PROTO_TIMEOUT_GRE_STREAM,
 #endif
+	NF_SYSCTL_CT_GC_SCAN_INTERVAL_MAX,
 
 	NF_SYSCTL_CT_LAST_SYSCTL,
 };
@@ -920,6 +921,14 @@ static struct ctl_table nf_ct_sysctl_table[] = {
 		.proc_handler   = proc_dointvec_jiffies,
 	},
 #endif
+	[NF_SYSCTL_CT_GC_SCAN_INTERVAL_MAX] = {
+		.procname	= "nf_conntrack_gc_scan_interval_max",
+		.data		= &nf_conntrack_gc_scan_interval_max,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_jiffies,
+		.extra1		= SYSCTL_ONE,
+	},
 };
 
 static struct ctl_table nf_ct_netfilter_table[] = {
@@ -1043,6 +1052,7 @@ static int nf_conntrack_standalone_init_sysctl(struct net *net)
 		table[NF_SYSCTL_CT_MAX].mode = 0444;
 		table[NF_SYSCTL_CT_EXPECT_MAX].mode = 0444;
 		table[NF_SYSCTL_CT_BUCKETS].mode = 0444;
+		table[NF_SYSCTL_CT_GC_SCAN_INTERVAL_MAX].mode = 0444;
 	}
 
 	cnet->sysctl_header = register_net_sysctl_sz(net, "net/netfilter",
-- 
2.50.1 (Apple Git-155)

Re: [PATCH net-next v2] netfilter: conntrack: expose gc_scan_interval_max via sysctl

[Florian Westphal]

Prasanna S Panchamukhi <panchamukhi@arista.com> wrote:
> The conntrack garbage collection worker uses an adaptive algorithm that
> adjusts the scan interval based on the average timeout of tracked
> entries.  The upper bound of this interval is hardcoded as
> GC_SCAN_INTERVAL_MAX (60 seconds).

I already said that I'm not keen on this approach.
Its a 'we can't do any better' type "solution".

If anything I'd be more inclined to make a change that allows to
more easily override the next_run computation via bpf.

Re: [PATCH net-next v2] netfilter: conntrack: expose gc_scan_interval_max via sysctl

[Pablo Neira Ayuso]

On Fri, Mar 13, 2026 at 03:09:19AM +0100, Florian Westphal wrote:
> Prasanna S Panchamukhi <panchamukhi@arista.com> wrote:
> > The conntrack garbage collection worker uses an adaptive algorithm that
> > adjusts the scan interval based on the average timeout of tracked
> > entries.  The upper bound of this interval is hardcoded as
> > GC_SCAN_INTERVAL_MAX (60 seconds).
> 
> I already said that I'm not keen on this approach.
> Its a 'we can't do any better' type "solution".
>
> If anything I'd be more inclined to make a change that allows to
> more easily override the next_run computation via bpf.

It is regrettable that the request for this knob appears to be
intended to enable a potentially proprietary hardware offload
extension, implemented through a userspace daemon and a proprietary
SDK.

It's 2026, there is plenty of infrastructure to offload the connection
tracking upstream, such as act_ct.c and the flowtable.

Re: [PATCH net-next v2] netfilter: conntrack: expose gc_scan_interval_max via sysctl

[Prasanna Panchamukhi]

On Fri, Mar 13, 2026 at 2:15 AM Pablo Neira Ayuso <pablo@netfilter.org> wrote:
>
> On Fri, Mar 13, 2026 at 03:09:19AM +0100, Florian Westphal wrote:
> > Prasanna S Panchamukhi <panchamukhi@arista.com> wrote:
> > > The conntrack garbage collection worker uses an adaptive algorithm that
> > > adjusts the scan interval based on the average timeout of tracked
> > > entries.  The upper bound of this interval is hardcoded as
> > > GC_SCAN_INTERVAL_MAX (60 seconds).
> >
> > I already said that I'm not keen on this approach.
> > Its a 'we can't do any better' type "solution".
> >
> > If anything I'd be more inclined to make a change that allows to
> > more easily override the next_run computation via bpf.
>
> It is regrettable that the request for this knob appears to be
> intended to enable a potentially proprietary hardware offload
> extension, implemented through a userspace daemon and a proprietary
> SDK.
>
> It's 2026, there is plenty of infrastructure to offload the connection
> tracking upstream, such as act_ct.c and the flowtable.

Thank you Pablo, for the suggestion. We will look into adopting the
hardware offload feature soon.

Thanks,
Prasanna