linux-ext4 - Re: jbd2 task hung in jbd2_journal_commit

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] [day] [month] [year] [list]

Message-ID: <CADve3d47X-q_+kn3=LhWV-D-FqtZ_9+juE1NqPwjAhRpVVwHjQ@mail.gmail.com>
Date:   Mon, 22 Mar 2021 12:04:19 +0530
From:   Shashidhar Patil <shashidhar.patil@...il.com>
To:     "Theodore Ts'o" <tytso@....edu>
Cc:     linux-ext4@...r.kernel.org
Subject: Re: jbd2 task hung in jbd2_journal_commit_transaction

Hi Theodore

On Fri, Mar 19, 2021 at 8:46 PM Theodore Ts'o <tytso@....edu> wrote:
>
> On Thu, Mar 18, 2021 at 12:27:44PM +0530, Shashidhar Patil wrote:
> > Hi Theodore,
> >     I forgot to include two important  details , the stack trace of
> > loop0 driver and sar output, which clearly nail  the problem.
> > The losetup with Ubuntu16.05 does not have O_DIRECT support  and we
> > were not aware of bypassing of journalling if
> > O_DIRECT combined with preallocated file scenario.
>
> Which version of the kernel are you using?  The use of O_DIRECT for

I am using *antediluvian* ubunut 16.04 LTS as below

$ uname -a
Linux ubuntu 4.15.0-117-generic #118~16.04.1-Ubuntu SMP Sat Sep 5
23:35:06 UTC 2020 x86_64 x86_64 x86_64 GNU/Linux

> loop devices requires kernel and losetup support.  (Also note that

As per the backtrace of loop0 driver attached it does look like a
problem, right ?

> upstream developers really generally don't pay attention --- or
> support --- distribution kernels unless they work for the company for
> which you are paying $$$ for support, and Ubuntu 16.05 isn't even a

$$$ again. No respite from $$ urge. $ Not good :-)

$ $$$
-bash: 31163$: command not found
:-)
> Long-Term Support distribution.)
>
> My suggestion is to see if you can replicate the problem on a modern
> userspace with the latest kernel.  And if not, then that's an object

I will try to recreate the problem with latest upstream kernel used by
upstream developers.

> lesson about why using a antediluvian is not a great life choice.  :-)
>
I agree with "antediluvian is not a great life choice" but can't
comment on "antediluvian is not a great choice in corporates".

Thank you
-Shashidhar
> Cheers,
>
>                                                 - Ted