linux-kernel - Re: [PATCH] scripts: prune-kernel:remove old kernels and modules dir from system

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: <b45d0312-d482-2d09-d6af-7a8b6b7051c5@infradead.org>
Date:   Thu, 31 Oct 2019 22:45:04 -0700
From:   Randy Dunlap <rdunlap@...radead.org>
To:     Bhaskar Chowdhury <unixbhaskar@...il.com>
Cc:     bfields@...ldses.org, yamada.masahiro@...ionext.com,
        michal.lkml@...kovi.net, linux-kbuild@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] scripts: prune-kernel:remove old kernels and modules dir
 from system

On 10/31/19 10:11 PM, Bhaskar Chowdhury wrote:
> On 09:53 Fri 01 Nov 2019, Bhaskar Chowdhury wrote:
>> On 08:06 Thu 31 Oct 2019, Randy Dunlap wrote:
>>> On 10/31/19 12:18 AM, Bhaskar Chowdhury wrote:
>>>> On 22:27 Wed 30 Oct 2019, Randy Dunlap wrote:
>>>>> On 10/30/19 9:52 PM, Bhaskar Chowdhury wrote:
>>>>>>>>> That 'rm' doesn't remove any files.  Compare what remove_old_kernel() does.
>>>>>>>> No,it is not using that function rather take the parameter from the
>>>>>>>> commandline and get into boot dir match with it and remove it.
>>>>>>>
>>>>>>> But it doesn't do that.  I tested it.  It should be more like what
>>>>>>> rmeove_old_kernel() does:
>>>>>>>
>>>>>>>         rm -If vmlinuz-$kernel_ver System.map-$kernel_ver config-$kernel_ver
>>>>>>>
>>>>>>> and if not, please explain why not.
>>>>>> Okay, again some uniformity missing in the code, I would like to your
>>>>>> suggested method,i.e call remove_old_kernel to do the job instead of depending on individual kernel.
>>>>>
>>>>> The simplest thing to do is set kernel_version=$kernel_ver
>>>>> and then call remove_old_kernel().
>>>>> And set modules_version=$modules_dir_name and call remove_old_modules_dir().
>>>>>
>>>>> But it would be cleaner to pass a parameter (kernel_version) to the
>>>>> remove_old_kernel() function and to pass a parameter (modules_dir) to the
>>>>> remove_old_modules_dir() function.
>>>>
>>>>>
>>>>
>>>> Thank you...I have just modified the code and call both the function
>>>> under remove option. BTW I didn't set the extra variable $kernel_ver name it $kernel_version and instead of $modules_dir_name name it $mo
>>>> dules_version.
>>>>
>>>> Capturing command line parameter in $kernel_version and $modules_version
>>>>
>>>> Is that fine? Here is a code snippet:
>>>
>>> Yes, that should be OK.
>>>
>>>> -r | --remove)
>>>>   if [[ $# -ne 3 ]]; then
>>>>    printf "You need to provide kernel version and modules directory
>>>>    name \n"
>>>>    exit 1
>>>>    else
>>>>    remove_old_kernel
>>>>    remove_old_modules_dir
>>>>    fi
>>>>
>>>>
>>>> I have just test it and it works.
>>>>
>>>>
>>>> And about solitary r option without hypen is ignoring and doing nothing.
>>>>
>>>> Means, if I pass ./scripts/prune-kernel r 5.3.3
>>>> it simply ignore and does nothing.Only with the hypen it can work.
>>>
>>> Is that how it should be?
>>> or what would you expect that to do?
>> Yes it should be. Any malformed parameter should be discarded,except
>> what explicitly given in code form,as help reminder.
>> OR
>> Do you want me to reminds user that they are missing thing to operate
>> correctly??
>>>
>>> -- 
>>> ~Randy
>> Bhaskar
>>>
> Randy,
> 
> I think we should show the help message , whenever user put a malformed
> parameter, say they put something like this :
> 
> ./scripts/prune-kernel f 5.3.3
> 
> It should show the help message to indicate what exactly need to pass
> with the script to get the desired result.
> 
> What do you think??

I agree.  Entering garbage junk should spit out help.

-- 
~Randy