netdev - Re: tcp_diag for all network namespaces?

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20241210020057.26127-1-kuniyu@amazon.com>
Date: Tue, 10 Dec 2024 11:00:57 +0900
From: Kuniyuki Iwashima <kuniyu@...zon.com>
To: <dave.seddon.ca@...il.com>
CC: <netdev@...r.kernel.org>, <kuniyu@...zon.com>
Subject: Re: tcp_diag for all network namespaces?

From: dave seddon <dave.seddon.ca@...il.com>
Date: Mon, 9 Dec 2024 11:24:18 -0800
> G'day,
> 
> Short
> Is there a way to extract tcp_diag socket data for all sockets from
> all network name spaces please?

I think there's no such interface.

I remember there was a similar request for TCP BPF iterator,
but now it's difficult because each netns could have its own
TCP hash table for established connections.

> 
> Background
> I've been using tcp_diag to dump out TCP socket performance every
> minute and then stream the data via Kafka and then into a Clickhouse
> database.  This is awesome for socket performance monitoring.
> 
> Kubernetes
> I'd like to adapt this solution to <somehow> allow monitoring of
> kubernetes clusters, so that it would be possible to monitor the
> socket performance of all pods.  Ideally, a single process could open
> a netlink socket into each network namespace, but currently that isn't
> possible.
> 
> Would it be crazy to add a new feature to the kernel to allow dumping
> all sockets from all name spaces?

Iterating netns in userspace is much simpler than in kernel that needs
iterating net_namespace_list under net_rwsem and remembering the last
netns with the refcount bumped.

> 
> Maybe I'm missing some other better option(s)?
> 
> Thanks in advance