Linux kernel -stable discussions
 help / color / mirror / Atom feed
From: Max Vozeler <max@vozeler.com>
To: Sasha Levin <sashal@kernel.org>, Chuck Lever <cel@kernel.org>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Chris Mason <mason@kernel.org>, Jeff Layton <jlayton@kernel.org>
Cc: linux-nfs@vger.kernel.org, stable@vger.kernel.org,
	regressions@lists.linux.dev
Subject: Re: [Regression linux-6.1.y]: NFSv4 server: sec=krb5p mount fails with EIO in 6.1.192
Date: Sat, 21 Nov 2026 17:01:03 +0100	[thread overview]
Message-ID: <awHAP3VW5BgSI3py@dp.vpn.hinterhof.net> (raw)
In-Reply-To: <2026-11-21-1-daily-reply-0001@kernel.org>

On Sat, Nov 21, 2026 at 10:27:29AM -0500, Sasha Levin wrote:
> > Reproduced with vanilla 6.1.191 and 6.1.192
> >
> > Good: 6.1.190
> > Bad: 6.1.191, 6.1.192
> 
> This looks like 82cf21374fb7 ("SUNRPC: Convert unwrap_priv_data() to use
> xdr_stream"), which went into 6.1.191 as a prerequisite of 2188232509f1

That matches my bisect result, it points to the same commit.

> ("SUNRPC: svcauth_gss: enforce krb5 token minimum length"). On 6.1,
> nfsd_dispatch() still re-initializes the decode stream from the start of
> head[0], so after the conversion a krb5p call is decoded starting at the length
> and sequence number words instead of the arguments. krb5i isn't affected
> because its path wasn't converted on 6.1.
>
> I've reverted both for the next 6.1 release and re-applied only the token
> length check on top of the old unwrap code.
> 
> Could you check that 6.1.192 with 2188232509f1 and then 82cf21374fb7 reverted
> fixes krb5p for you?

It does indeed. 6.1.192 with those two reverted works fine.

Thanks!
Max

      reply	other threads:[~2026-11-21 16:01 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-20 13:17 [Regression linux-6.1.y]: NFSv4 server: sec=krb5p mount fails with EIO in 6.1.192 Max Vozeler
2026-11-21 10:28 ` Max Vozeler
2026-11-21 12:30   ` Max Vozeler
2026-11-21 12:54     ` Max Vozeler
2026-11-21 15:27   ` Sasha Levin
2026-11-21 16:01     ` Max Vozeler [this message]

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=awHAP3VW5BgSI3py@dp.vpn.hinterhof.net \
    --to=max@vozeler.com \
    --cc=cel@kernel.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=jlayton@kernel.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=mason@kernel.org \
    --cc=regressions@lists.linux.dev \
    --cc=sashal@kernel.org \
    --cc=stable@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
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