From: Chuck Lever <chuck.lever@oracle.com>
To: John <therealgraysky@proton.me>
Cc: "linux-nfs@vger.kernel.org" <linux-nfs@vger.kernel.org>
Subject: Re: kernel panic when starting nfsd on OpenWrt with kernel 6.12.19
Date: Sat, 22 Mar 2025 14:06:27 -0400 [thread overview]
Message-ID: <75ec06ad-7472-4e83-a20e-28543ec2909d@oracle.com> (raw)
In-Reply-To: <yW5ewBN3-dAMHSq9KmbFRPRt_fK0FTmuqclUbu4K1kZPcfB6DmXRPOVC_OAwh1waQz2Of0qUDqxQ3YL-NBULF9H6-HMdVjixFAad20f5sUY=@proton.me>
On 3/22/25 1:20 PM, John wrote:
>
>
> On Saturday, March 22nd, 2025 at 10:49 AM, Chuck Lever <chuck.lever@oracle.com> wrote:
>>
>> I recommend bisecting between 6.6 and 6.12.19 to locate the culprit.
>>
>> When you have that information, open a bug on bugzilla.kernel.org under
>> Filesystems/NFSD and attach your information there.
>
> Wish I could, but the way OpenWrt is designed, moving from one kernel branch to another is major undertaking often requiring months of tweaks 1,000s of patches, Makefiles, config files, etc.
>
> I have been reading up on NFSv4 and am wondering if there is a way to simply disable the callback function. Could that work-around this crash?
>
> Unfortunately, OpenWrt's nfs-kernel-server package does not ship nor use something like /etc/nfs.conf. At the link below you can see their init script to start nfsd. Many thank for any ideas.
>
> https://github.com/openwrt/packages/blob/master/net/nfs-kernel-server/files/nfsd.init
>
If this is a "product" kernel then you should approach the distributor
first. I do not have the hardware, the binaries, the .config, or a
cross compiler for whatever "Hardware name: iKOOLCORE R2Max/R2Max" is
(guessing RPi ?)
It would be helpful to get
$ scripts/faddr2line fs/nfsd/nfs4callback.c \
nfsd4_probe_callback_sync+0x1094
although, IME, +1094 means the crash happened far outside of the
nfs4callback.c source file.
This:
nfsd4_client_tracking_init+0x39/0x150 [nfsd]
suggests that your distributor (or LTS) might be missing a fix
for recent crashes that are due to incorrectly setting
CONFIG_NFSD_LEGACY_CLIENT_TRACKING. I'm thinking of perhaps
de71d4e211ed ("nfsd: fix legacy client tracking initialization")
in particular, which does not appear to have been applied to
v6.12.19.
--
Chuck Lever
next prev parent reply other threads:[~2025-03-22 18:06 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-22 10:05 kernel panic when starting nfsd on OpenWrt with kernel 6.12.19 John
2025-03-22 14:49 ` Chuck Lever
2025-03-22 17:20 ` John
2025-03-22 17:33 ` Chuck Lever
2025-03-22 18:06 ` Chuck Lever [this message]
2025-03-22 18:52 ` John
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=75ec06ad-7472-4e83-a20e-28543ec2909d@oracle.com \
--to=chuck.lever@oracle.com \
--cc=linux-nfs@vger.kernel.org \
--cc=therealgraysky@proton.me \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.