Hi!

I have some tables and playing with a query and graphical explain got me
to a strange result - a grey line across the entire window; see
http://mcajvar.prkoritu.net/pgadmin/...xplain_db1.png .

I then made an exact duplicate copy of the users, database, everything I
could think of and tried to duplicate the strange display. But the
problem wouldn't occur in this local copy; see
http://mcajvar.prkoritu.net/pgadmin/...xplain_db2.png .

Does pgAdmin use the output of the explain query in order to construct
the graphic illustration? If so then these may be helpful:

This yields foo graphics:
Nested Loop (cost=0.00..0.01 rows=1 width=485)
Join Filter: ("outer".id = "inner".id_eventi)
-> Seq Scan on eventi t1 (cost=0.00..0.00 rows=1 width=12)
Filter: ((datum >= '2006-04-12 00:00:00'::timestamp without
time zone) AND (datum <= '2006-05-12 00:00:00'::timestamp without time
zone) AND (prikaz = 1::smallint))
-> Seq Scan on eventi_vsebina t2 (cost=0.00..0.00 rows=1 width=477)
Filter: (narocilnica = 1::smallint)

This yields a nice illustration:
Nested Loop (cost=0.00..47.26 rows=1 width=485)
Join Filter: ("outer".id = "inner".id_eventi)
-> Seq Scan on eventi t1 (cost=0.00..35.38 rows=1 width=12)
Filter: ((datum >= '2006-04-12 00:00:00'::timestamp without
time zone) AND (datum <= '2006-05-12 00:00:00'::timestamp without time
zone) AND (prikaz = 1::smallint))
-> Seq Scan on eventi_vsebina t2 (cost=0.00..11.88 rows=1 width=477)
Filter: (narocilnica = 1::smallint)

I tried refreshing multiple times in both databases, the results were
always the same. The foo display happens when working on a remote
database on Server 1, while the local tests were done on Server 2:
Server 1: 8.0.1 on DragonFlyBSD 1.2.6, compiled by GCC 2.95.4
Server 2: 8.0.3 on SuSE Linux 10.0, compiled by GCC 4.0.2

If there is any other info I can provide in order to help diagnose the
problem, let me know.

Regards,
Miha


---------------------------(end of broadcast)---------------------------
TIP 2: Don't 'kill -9' the postmaster

Miha Radej wrote:
> Hi!
>
> I have some tables and playing with a query and graphical explain got me
> to a strange result - a grey line across the entire window; see
> http://mcajvar.prkoritu.net/pgadmin/...xplain_db1.png .

There's something wrong here :-)

>
> I then made an exact duplicate copy of the users, database, everything I
> could think of and tried to duplicate the strange display. But the
> problem wouldn't occur in this local copy; see
> http://mcajvar.prkoritu.net/pgadmin/...xplain_db2.png .
>
> Does pgAdmin use the output of the explain query in order to construct
> the graphic illustration?

It does.

If so then these may be helpful:
>
> This yields foo graphics:
> Nested Loop (cost=0.00..0.01 rows=1 width=485)
> Join Filter: ("outer".id = "inner".id_eventi)
> -> Seq Scan on eventi t1 (cost=0.00..0.00 rows=1 width=12)
> Filter: ((datum >= '2006-04-12 00:00:00'::timestamp without
> time zone) AND (datum <= '2006-05-12 00:00:00'::timestamp without time
> zone) AND (prikaz = 1::smallint))
> -> Seq Scan on eventi_vsebina t2 (cost=0.00..0.00 rows=1 width=477)
> Filter: (narocilnica = 1::smallint)
>
> This yields a nice illustration:
> Nested Loop (cost=0.00..47.26 rows=1 width=485)
> Join Filter: ("outer".id = "inner".id_eventi)
> -> Seq Scan on eventi t1 (cost=0.00..35.38 rows=1 width=12)
> Filter: ((datum >= '2006-04-12 00:00:00'::timestamp without
> time zone) AND (datum <= '2006-05-12 00:00:00'::timestamp without time
> zone) AND (prikaz = 1::smallint))
> -> Seq Scan on eventi_vsebina t2 (cost=0.00..11.88 rows=1 width=477)
> Filter: (narocilnica = 1::smallint)

Hm, don't see why this clashes, something wrong with cost=0? Will have
to debug this some day.

Regards,
Andreas

---------------------------(end of broadcast)---------------------------
TIP 6: explain analyze is your friend

Miha Radej wrote:
> Hi!
>
> I have some tables and playing with a query and graphical explain got me
> to a strange result - a grey line across the entire window; see
> http://mcajvar.prkoritu.net/pgadmin/...xplain_db1.png .

Ok Miha,

apparently the problem is triggered by multiline filters. I added an
option to pgAdmin (svn) that allows graphical display of a loaded
analyze output, the multiline issue is fixed now.

Please try svn.

Regards,
Andreas

---------------------------(end of broadcast)---------------------------
TIP 4: Have you searched our list archives?

http://archives.postgresql.org