postgres-xc-committers Mailing List for Postgres-XC

Thanks Michael;

This explains well why we had errors after a while.   I have only one question:

I believe that vacuum should remove main tuple, then TOASTed tuples,
to avoid dangling TOAST reference.

Maybe we need to ask Jan for this.   Because vacuumed tuples are
invisible, order of vacuum may not be an issue at all.   It is simpler
to vacuum referred TOAST first.

Regards;
----------
Koichi Suzuki


2012/5/24 Michael Paquier <mic...@us...>:
> Project "Postgres-XC".
>
> The branch, REL1_0_STABLE has been updated
>       via  c9a15eeaaa78cf47f57830bbcb3234917e041afa (commit)
>      from  fcdf4ebba6613c7e7de2631325cf1ce682c684e4 (commit)
>
>
> - Log -----------------------------------------------------------------
> http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=c9a15eeaaa78cf47f57830bbcb3234917e041afa
>
> commit 39330d67e4c6a6a3ce8fe3abea1d60ca645ed1ff
> Author: Michael Paquier <mi...@ot...>
> Date:   Thu May 24 17:32:56 2012 +0900
>
>    Fix for bug 3525691: autovacuum process issue
>
>    This commit fixes 2 issues:
>    1) On Coordinator, autovacuum processes were using local snapshots
>    while it needs to fetch global snapshot data from GTM. This made several
>    PostgreSQL internal processes going mad with visibility like statistics or
>    catalogs in an environment using a lot of DDL.
>    2) On Datanodes, autovacuum non-analyze processes did not fetch a global
>    transaction ID from GTM and this transaction ID was not included in global
>    snapshot. Hence, global data consistency and visibility was compromised
>    on local nodes. This issue made autovacuum and pg_toast going crazy if the
>    system was put under heavy load (transaction timeout occuring).
>
>    It is believed that this bug is here since Postgres-XC 0.9.4, far before the
>    code was merged with PostgreSQL 9.1, and was causing numerous issues with
>    long and short-time runs.
>    For a reason I do not completely understand, this bug has become really easy
>    to reproduce since some race conditions in pg_toast code (tuptoaster.c) have
>    been fixed by Tom Lane in 9.1 stable.
>
> M       src/backend/access/transam/varsup.c
> M       src/backend/storage/ipc/procarray.c
>
> -----------------------------------------------------------------------
>
> Summary of changes:
>  src/backend/access/transam/varsup.c |   14 +++++++++-----
>  src/backend/storage/ipc/procarray.c |   16 +++++++++-------
>  2 files changed, 18 insertions(+), 12 deletions(-)
>
>
> hooks/post-receive
> --
> Postgres-XC
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> Postgres-xc-committers mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-committers

Project "Postgres-XC".

The branch, master has been updated
       via  39330d67e4c6a6a3ce8fe3abea1d60ca645ed1ff (commit)
      from  e1df051ccd64fa9eee5562f8a64be7007003120d (commit)


- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=39330d67e4c6a6a3ce8fe3abea1d60ca645ed1ff

commit 39330d67e4c6a6a3ce8fe3abea1d60ca645ed1ff
Author: Michael Paquier <mi...@ot...>
Date:   Thu May 24 17:32:56 2012 +0900

    Fix for bug 3525691: autovacuum process issue
    
    This commit fixes 2 issues:
    1) On Coordinator, autovacuum processes were using local snapshots
    while it needs to fetch global snapshot data from GTM. This made several
    PostgreSQL internal processes going mad with visibility like statistics or
    catalogs in an environment using a lot of DDL.
    2) On Datanodes, autovacuum non-analyze processes did not fetch a global
    transaction ID from GTM and this transaction ID was not included in global
    snapshot. Hence, global data consistency and visibility was compromised
    on local nodes. This issue made autovacuum and pg_toast going crazy if the
    system was put under heavy load (transaction timeout occuring).
    
    It is believed that this bug is here since Postgres-XC 0.9.4, far before the
    code was merged with PostgreSQL 9.1, and was causing numerous issues with
    long and short-time runs.
    For a reason I do not completely understand, this bug has become really easy
    to reproduce since some race conditions in pg_toast code (tuptoaster.c) have
    been fixed by Tom Lane in 9.1 stable.

M	src/backend/access/transam/varsup.c
M	src/backend/storage/ipc/procarray.c

-----------------------------------------------------------------------

Summary of changes:
 src/backend/access/transam/varsup.c |   14 +++++++++-----
 src/backend/storage/ipc/procarray.c |   16 +++++++++-------
 2 files changed, 18 insertions(+), 12 deletions(-)


hooks/post-receive
-- 
Postgres-XC

Project "Postgres-XC".

The branch, REL1_0_STABLE has been updated
       via  c9a15eeaaa78cf47f57830bbcb3234917e041afa (commit)
      from  fcdf4ebba6613c7e7de2631325cf1ce682c684e4 (commit)


- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=c9a15eeaaa78cf47f57830bbcb3234917e041afa

commit 39330d67e4c6a6a3ce8fe3abea1d60ca645ed1ff
Author: Michael Paquier <mi...@ot...>
Date:   Thu May 24 17:32:56 2012 +0900

    Fix for bug 3525691: autovacuum process issue
    
    This commit fixes 2 issues:
    1) On Coordinator, autovacuum processes were using local snapshots
    while it needs to fetch global snapshot data from GTM. This made several
    PostgreSQL internal processes going mad with visibility like statistics or
    catalogs in an environment using a lot of DDL.
    2) On Datanodes, autovacuum non-analyze processes did not fetch a global
    transaction ID from GTM and this transaction ID was not included in global
    snapshot. Hence, global data consistency and visibility was compromised
    on local nodes. This issue made autovacuum and pg_toast going crazy if the
    system was put under heavy load (transaction timeout occuring).
    
    It is believed that this bug is here since Postgres-XC 0.9.4, far before the
    code was merged with PostgreSQL 9.1, and was causing numerous issues with
    long and short-time runs.
    For a reason I do not completely understand, this bug has become really easy
    to reproduce since some race conditions in pg_toast code (tuptoaster.c) have
    been fixed by Tom Lane in 9.1 stable.

M	src/backend/access/transam/varsup.c
M	src/backend/storage/ipc/procarray.c

-----------------------------------------------------------------------

Summary of changes:
 src/backend/access/transam/varsup.c |   14 +++++++++-----
 src/backend/storage/ipc/procarray.c |   16 +++++++++-------
 2 files changed, 18 insertions(+), 12 deletions(-)


hooks/post-receive
-- 
Postgres-XC