Re: [v9.1] sepgsql - userspace access vector cache - Mailing list pgsql-hackers

From Kohei Kaigai
Subject Re: [v9.1] sepgsql - userspace access vector cache
Date
Msg-id D0C1A1F8BF513F469926E6C71461D9EC01DB31@EX10MBX02.EU.NEC.COM
Whole thread Raw
In response to Re: [v9.1] sepgsql - userspace access vector cache  (Robert Haas <robertmhaas@gmail.com>)
List pgsql-hackers
> On Wed, Jul 20, 2011 at 12:04 PM, Kohei Kaigai
> <Kohei.Kaigai@emea.nec.com> wrote:
> > The sepgsql_restorecon(NULL) assigns default security label on all the
> > database objects being controlled, thus, its workload caches security
> > label (including text data) of these objects.
> > So, ~5MB of difference is an upper limit of syscache usage because of
> > SECLABELOID.
>
> No, it's not.  It's just the upper limit of how large it can be on an
> *empty* database.  A real database could have hundreds of tables and
> views and thousands of columns.  To say nothing of large objects.
>
Ah, sorry, you are correct.

Regarding to large objects, GetSecurityLabel() is modified not to use
SECLABELOID to flood of the syscache.

Thanks,
--
NEC Europe Ltd, SAP Global Competence Center
KaiGai Kohei <kohei.kaigai@emea.nec.com>


pgsql-hackers by date:

Previous
From: Kohei Kaigai
Date:
Subject: Re: [v9.1] sepgsql - userspace access vector cache
Next
From: Tom Lane
Date:
Subject: Re: Another issue with invalid XML values