Hello,
To bypass the 512 bytes limitation of a /etc/group line I usually used
the following :
group::30:<members>
group1::30:<members>

Until now it works for all what I need.
But It doesn't work with our DB2 db.
DB2 can not see a user that is part of the unix group "group1"

So what can be the consequence if I use the following bypass :

group::30:<members>
group::30:<members>

same GID but also same group name

Thank you

Kona wrote:

> So what can be the consequence if I use the following bypass :
> 
> group::30:<members>
> group::30:<members>
> 
> same GID but also same group name

We've used that format for years without any problems.
Of course, YMMV.

kona_iron@yahoo.fr (Kona) writes:
>To bypass the 512 bytes limitation of a /etc/group line I usually used
>the following :
>group::30:<members>
>group1::30:<members>

I only know of a 1024 byte limit, and that only applies to 
yp and is related to ndb. Is there another limitation that I don't
know about?

-Mike "no answers, only questions..."

hubcap wrote:
> kona_iron@yahoo.fr (Kona) writes:
> 
>>To bypass the 512 bytes limitation of a /etc/group line I usually used
>>the following :
>>group::30:<members>
>>group1::30:<members>
> 
> I only know of a 1024 byte limit, and that only applies to 
> yp and is related to ndb. Is there another limitation that I don't
> know about?


Some programs seem to have 512 hardcoded (bug?)

# uname -a
SunOS hostname 5.10 s10_69 sun4u sparc SUNW,Sun-Blade-100

# strings /usr/sbin/useradd | grep Group
WARNING: Group entry exceeds 512 char: /etc/group entry truncated.