From: Pablo Neira Ayuso <pablo@netfilter•org>
To: Prasanna Panchamukhi <panchamukhi@arista•com>
Cc: Florian Westphal <fw@strlen•de>,
netfilter-devel@vger•kernel.org,
"David S. Miller" <davem@davemloft•net>,
Eric Dumazet <edumazet@google•com>,
Jakub Kicinski <kuba@kernel•org>, Paolo Abeni <pabeni@redhat•com>,
Simon Horman <horms@kernel•org>, Jonathan Corbet <corbet@lwn•net>,
Shuah Khan <skhan@linuxfoundation•org>, Phil Sutter <phil@nwl•cc>,
netdev@vger•kernel.org, linux-doc@vger•kernel.org,
linux-kernel@vger•kernel.org, coreteam@netfilter•org
Subject: Re: [PATCH net-next] netfilter: conntrack: expose gc_scan_interval_max via sysctl
Date: Thu, 12 Mar 2026 23:42:36 +0100 [thread overview]
Message-ID: <abNBXCgi4x5WLkBa@chamomile> (raw)
In-Reply-To: <CACqWiXD2_O32K4NhmNBZrAUG7U9-N93LTFjJHG6Tq=4vuafNuA@mail.gmail.com>
Hi Prasanna,
On Thu, Mar 12, 2026 at 03:31:06PM -0700, Prasanna Panchamukhi wrote:
> On Thu, Mar 12, 2026 at 5:36 AM Florian Westphal <fw@strlen•de> 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).
> > >
> > > 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.
> >
> > This was proposed before, see:
> >
> > https://lore.kernel.org/netfilter-devel/aO-id5W6Tr7frdHN@strlen.de/
> > https://lore.kernel.org/netfilter-devel/aRsuU57juCvsMBKE@strlen.de/
> >
> > I did not hear back wrt. the horizon cache.
> >
> > I'm not 100% opposed to this, but I do wonder if we really can't do
> > better than the current avg strategy.
>
> Hi Florian,
>
> Our primary goal is to cap the maximum time taken by the GC to clean
> up expired entries. We rely on user-space notifications to clean up
> these entries from the hardware, so ensuring a predictable upper bound
> is important for our use case.
Is there any reason why you decide not to use instead the existing
hardware offload infrastructure for this purpose?
next prev parent reply other threads:[~2026-03-12 22:42 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-11 19:40 [PATCH net-next] netfilter: conntrack: expose gc_scan_interval_max via sysctl Prasanna S Panchamukhi
2026-03-12 12:12 ` Fernando Fernandez Mancera
2026-03-12 12:15 ` Fernando Fernandez Mancera
2026-03-12 21:44 ` Prasanna Panchamukhi
2026-03-12 12:36 ` Florian Westphal
2026-03-12 22:31 ` Prasanna Panchamukhi
2026-03-12 22:42 ` Pablo Neira Ayuso [this message]
2026-03-12 23:10 ` Florian Westphal
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=abNBXCgi4x5WLkBa@chamomile \
--to=pablo@netfilter$(echo .)org \
--cc=corbet@lwn$(echo .)net \
--cc=coreteam@netfilter$(echo .)org \
--cc=davem@davemloft$(echo .)net \
--cc=edumazet@google$(echo .)com \
--cc=fw@strlen$(echo .)de \
--cc=horms@kernel$(echo .)org \
--cc=kuba@kernel$(echo .)org \
--cc=linux-doc@vger$(echo .)kernel.org \
--cc=linux-kernel@vger$(echo .)kernel.org \
--cc=netdev@vger$(echo .)kernel.org \
--cc=netfilter-devel@vger$(echo .)kernel.org \
--cc=pabeni@redhat$(echo .)com \
--cc=panchamukhi@arista$(echo .)com \
--cc=phil@nwl$(echo .)cc \
--cc=skhan@linuxfoundation$(echo .)org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox