Обсуждение: Re: [GENERAL] have trouble understanding xmin and xmax with updateoperations from two different sessions

have trouble understanding xmin and xmax with update operations from two
different sessions

So, as found below, Session2 is trying to update a row which is already
updated with a different value and it's update fails with *UPDATE 0*

But from Session3, I see that xmax value is visible as Session2's txid. Why
is it like that?
Can we not put Session2's txid to xmin instead(although the update failed)?
And if we try to consider that xmax is update with Session2's txid bcoz the
update failed, then why bother updating the xmax?

Please help me understand this.

*Session1*
testdb=# BEGIN;
BEGIN
testdb=# select ctid, xmin, xmax, * from numbers;
ctid  |  xmin  | xmax | number
-------+--------+------+--------
(0,2) | 519107 |    0 |     14
(0,3) | 519112 |    0 |     23
(2 rows)

testdb=# select txid_current();
txid_current
--------------
       519114
(1 row)

testdb=# update numbers set number=24 where number=14;
UPDATE 1
testdb=# COMMIT;
COMMIT


*Session 2*
testdb=# BEGIN;
BEGIN
testdb=# select txid_current();
txid_current
--------------
       519115
(1 row)

testdb=# select ctid, xmin, xmax, * from numbers;
ctid  |  xmin  | xmax | number
-------+--------+------+--------
(0,2) | 519107 |    0 |     14
(0,3) | 519112 |    0 |     23
(2 rows)

testdb=# select ctid, xmin, xmax, * from numbers;
ctid  |  xmin  |  xmax  | number
-------+--------+--------+--------
(0,2) | 519107 | 519114 |     14
(0,3) | 519112 |      0 |     23
(2 rows)

testdb=# update numbers set number=25 where number=14;
UPDATE 0
testdb=# COMMIT;
COMMIT

*Session 3*
testdb=# select txid_current();
txid_current
--------------
       519116
(1 row)

testdb=# select ctid, xmin, xmax, * from numbers;
ctid  |  xmin  |  xmax  | number
-------+--------+--------+--------
(0,3) | 519112 |      0 |     23
(0,4) | 519114 | 519115 |     24
(2 rows)




-----
--
Thanks,
Rajan.
--
View this message in context: http://www.postgresql-archive.org/have-trouble-understanding-xmin-and-xmax-with-update-operations-from-two-different-sessions-tp5969629.html
Sent from the PostgreSQL - general mailing list archive at Nabble.com.


--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

hello,

thanks for replies, Adrian, Steven.

>So calling it can advance the xid manually. Some testing here showed
>that what xmin or xmax is created depends on when you call txid_current
>in either the original session or the concurrent sessions.

I understand this and I am executing my statements inside a Transaction
block so the xid is not incremented when calling it.

>Also worth noting that an UPDATE in Postgres is a DELETE/INSERT process.
>The clue is the ctid value. In  Session 2 you are looking at the
>original row(ctid=(0, 2) which has been marked as deleted(non-zero
>xmax). In Session 3 you are looking at the new row(ctid(0, 4)).

Yes. But why (ctid(0,4)) in *Session 3* carries the xmax of the txid 519115
in which the update failed with *UPDATE 0* . This is where I can not
understand,
1. Row (0,4) is updated with correct value and (0,3) is not visible in
Session 2, which is good.
2. but in *Session 3* (0,4) also carries xmax which means what? Is it also
marked for deletion? It can't be, right?




-----
--
Thanks,
Rajan.
--
View this message in context:
http://www.postgresql-archive.org/Re-have-trouble-understanding-xmin-and-xmax-with-update-operations-from-two-different-sessions-tp5969644p5969656.html
Sent from the PostgreSQL - general mailing list archive at Nabble.com.

Thanks, Jeff. That helps understanding it 50%.

*Session 2* fails to UPDATE the record which is in *(0,2)* and this tuple is
marked for deletion. It means that *(0,2) never exists* when Session 2 is
trying to perform the update.

In that case, how *Session 3's new row (0,4)* contains the xmax as *Session
2's txid*.



-----
--
Thanks,
Rajan.
--
View this message in context:
http://www.postgresql-archive.org/Re-have-trouble-understanding-xmin-and-xmax-with-update-operations-from-two-different-sessions-tp5969644p5969661.html
Sent from the PostgreSQL - general mailing list archive at Nabble.com.

Thanks for the explanation.

will I be able to view the information using this function,
SELECT * FROM heap_page_items(get_raw_page('testnumbers', 0));

Also, please let me know which column I should refer for viewing the
pointer.



-----
--
Thanks,
Rajan.
--
View this message in context:
http://www.postgresql-archive.org/Re-have-trouble-understanding-xmin-and-xmax-with-update-operations-from-two-different-sessions-tp5969644p5969767.html
Sent from the PostgreSQL - general mailing list archive at Nabble.com.

Thanks, Jeff.

Now I am going back to my old question.

Even though *Session 2* fails to update with UPDATE 0 message, its txid is
saved in xmax of updated(by *Session 1*) tuple.

As it becomes an old txid, how come new txids are able to view it?



-----
--
Thanks,
Rajan.
--
View this message in context:
http://www.postgresql-archive.org/Re-have-trouble-understanding-xmin-and-xmax-with-update-operations-from-two-different-sessions-tp5969644p5969857.html
Sent from the PostgreSQL - general mailing list archive at Nabble.com.