Re: [Bacula-devel] Major problem in restores

[Kern S. <ke...@si...>]

Hello,

On Fri, 2004-03-26 at 13:23, Dan Langille wrote:
> On Fri, 26 Mar 2004, Kern Sibbald wrote:
> 
> > On Thu, 2004-03-25 at 23:27, Dan Langille wrote:
> > > On Thu, 25 Mar 2004, Phil Stracchino wrote:
> > >
> > > > > Currently, Bacula never removes Filename entries nor Path entries since
> > > > > the overhead of finding out if they are orphaned is too high. Long term,
> > > > > this needs a solution.
> > >
> > > I may look at a solution for PostgreSQL.  This should be a trivial thing
> > > do to.  In fact, it may already be implemented.
> >
> > I'll be interested in seeing the solution you find.
> 
> My time until mid-May is pretty much absorbed by BSDCan.

No problems.

> 
> > > > Would adding a reference count to Filename and Path records be a viable
> > > > way to handle the problem?  Increment the reference count each time a
> > > > File record uses it, decrement each time a File record referencing it is
> > > > pruned.  (For safety, the refcount would probably have to be a bigint.)
> > > > Or would that add an unacceptable amount of overhead?
> > >
> > > Phil: That's a job for the database.  That's what Foreign Keys and Primary
> > > Keys for for.  Yes, I know MySQL doesn't do that very well yet.  But
> > > PostgreSQL does it very well.  It's in that option.
> >
> > Unfortunately, SQLite doesn't worry about it either, so despite the
> > magic you work with PostgreSQL, I'll need to do something in dbcheck ...
> 
> Does SQLite support nested queries?

Quite likely, but MySQL does not, though I understand that they do exist
in at least some version of 4.0. Many people such as myself are still
using 3.xx.

> 
> test=# create table a (id int);
> CREATE TABLE
> test=# create table b (id int);
> CREATE TABLE
> test=# insert into a values (1);
> INSERT 26936 1
> test=# insert into a values (Segmentation fault (core dumped)
> [dan@laptop:~] $
> [dan@laptop:~] $ rm psql.core
> [dan@laptop:~] $ psql test
> Welcome to psql 7.4.2, the PostgreSQL interactive terminal.
> 
> Type:  \copyright for distribution terms
>        \h for help with SQL commands
>        \? for help on internal slash commands
>        \g or terminate with semicolon to execute query
>        \q to quit
> 
> create table a (id int);
> create table b (id int);
> insert into a values (1);
> insert into a values (2);
> insert into b values (2);
> select * from a;
>  id
> ----
>   1
>   2
> (2 rows)
> 
> select * from  b;
>  id
> ----
>   2
> (1 row)
> 
> select * from b where exists (select * from a where a.id =  b.id);
>  id
> ----
>   2
> (1 row)
> 
> If that works, it makes your clean up much easier.