[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <f0abd655-0c10-4190-97bf-34ffb92f98ad@kernel.org>
Date: Fri, 12 Jan 2024 19:39:20 +0100
From: Matthieu Baerts <matttbe@...nel.org>
To: Quentin Deslandes <qde@...cy.de>, netdev@...r.kernel.org
Cc: David Ahern <dsahern@...il.com>,
Stephen Hemminger <stephen@...workplumber.org>
Subject: Re: [PATCH] ss: prevent "Process" column from being printed unless
requested
Hi Quentin,
On 06/12/2023 12:14, Quentin Deslandes wrote:
> Commit 5883c6eba517 ("ss: show header for --processes/-p") added
> "Process" to the list of columns printed by ss. However, the "Process"
> header is now printed even if --processes/-p is not used.
>
> This change aims to fix this by moving the COL_PROC column ID to the same
> index as the corresponding column structure in the columns array, and
> enabling it if --processes/-p is used.
I recently noticed that some MPTCP selftests were broken after having
upgraded IPRoute2 to the latest version: from v6.6 to v6.7.
It looks like this is due to this patch here:
1607bf53 ("ss: prevent "Process" column from being printed unless
requested")
If I revert it, MPTCP selftests are passing again.
>From what I can see, the behaviour has changed recently: if '-i' is used
without '-p', the extra line is no longer not displayed any more, e.g.
'ss -tl' and 'ss -tli' gives the same output:
> $ ss -tl
> State Recv-Q Send-Q Local Address:Port Peer Address:Port
> LISTEN 0 1 0.0.0.0:4444 0.0.0.0:*
> $ ss -tli
> State Recv-Q Send-Q Local Address:Port Peer Address:Port
> LISTEN 0 1 0.0.0.0:4444 0.0.0.0:*
If '-p' is added, '-i' works as before:
> $ ss -tlp
> State Recv-Q Send-Q Local Address:Port Peer Address:Port Process
> LISTEN 0 1 0.0.0.0:4444 0.0.0.0:* users:(("netcat",pid=2668,fd=3))
> $ ss -tlpi
> State Recv-Q Send-Q Local Address:Port Peer Address:Port Process
> LISTEN 0 1 0.0.0.0:4444 0.0.0.0:* users:(("netcat",pid=2668,fd=3))
> cubic cwnd:10
(we can see the extra line with 'cubic cwnd:10')
I just noticed that you were working on other patches modifying 'ss',
and even adding a new column. Are you aware of this issue? Are you
already working -- or going to work -- on a fix for it?
Cheers,
Matt
--
Sponsored by the NGI0 Core fund.
Powered by blists - more mailing lists