Openembedded Core Discussions
 help / color / mirror / Atom feed
From: Martin Donnelly <martin.donnelly@ge.com>
To: Mark Hatle <mark.hatle@windriver.com>
Cc: openembedded-core@lists.openembedded.org
Subject: Re: [PATCH] cracklib: generate the default dictionary on the target
Date: Thu, 25 Apr 2013 16:37:47 +0100	[thread overview]
Message-ID: <51794DCB.5020603@ge.com> (raw)
In-Reply-To: <51793CE0.7060809@windriver.com>

On 25/04/2013 15:25, Mark Hatle wrote:
> On 4/25/13 8:28 AM, Martin Donnelly wrote:
>> Currently cracklib-native is used to generate the dictionary
>> database from the wordlist during install. Unfortunately the
>> database files are endian-specific and this results in
>> errors when the host and target endianness differ.
>>
>> This patch creates the default dictionary files on the target
>> during post-install. In order to do this the default wordlist has
>> also been compressed, this enables an unmodified cracklib-format
>> to correctly parse the input file with both busybox and GNU gzip.
>> (These behave differently if the input is uncompressed, GNU
>> gzip behaves like cat while busybox returns an error).
>>
>> Finally the cracklib-native package has been removed since
>> it is no longer used.
>
> I thought we had fixed cracklib to have a fixed endian.  If not, this is
> a bug.  We want to avoid running things like the cracklib generation on
> the target if possible, as it impacts the first boot time and read-only
> roots.
>

I had a look for any patches which fixed this but couldn't see any, 
there were some related to using dictionaries between 32 and 64-bit 
machines of the same endianness but I maybe missed another one? I wasn't 
100% happy sending it for the issues you mention but thought it better 
to post it than not.

> (I'm fine with this as a workaround, but it's not the fix I'd like.)
> I'd be happy if we simply fixed the cracklib-native to pass in an endian
> and have the right output generated.  (Note, we also have to consider
> structure alignment/packing as well.)
>

Agreed, this would be best and it's something I'll look at when I've got 
the time.

> If there isn't already a bug in the Yocto Project bugzilla, can you add
> one?
>

Done, https://bugzilla.yoctoproject.org/show_bug.cgi?id=4419

-Martin



      reply	other threads:[~2013-04-25 15:55 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-25 13:28 [PATCH] cracklib: generate the default dictionary on the target Martin Donnelly
2013-04-25 14:25 ` Mark Hatle
2013-04-25 15:37   ` Martin Donnelly [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=51794DCB.5020603@ge.com \
    --to=martin.donnelly@ge.com \
    --cc=mark.hatle@windriver.com \
    --cc=openembedded-core@lists.openembedded.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