Hello,
two last questions about the transaction ID wraparound problem of my DBs I had
yesterday (see former postings). After applying the 'recovery procedure' Tom
suggested now my XIDs are looking almost fine again, except for the template0
DB.
dev_db=# SELECT datname, age(datfrozenxid) FROM pg_database;
datname | age
-------------+-------------
xgc | 1075434814
dev_db | 1074296718
template1 | 1075486644
template0 | -1882457315
promo_db | 1075528357
snap_db_new | 1075528467
gendev_db | 1075512627
Is this negligible or can it cause any harm in the future? Is there a way to
VACUUM template0 as well?
My second point is more a suggestion. After Toms strategy worked quite well (at
least for me) it is maybe worthwhile to put it somewhere in the docs. Maybe
together with other things in a section called 'Disaster recovery strategies'?
regards
Michael
--
------------------------------------------------------------------------------
Dipl.-Inform. Michael Beckstette Office: M3-129
AG-PI / Technische Fakultaet EMail:[EMAIL PROTECTED]
Universitaet Bielefeld Fon: +49-521-106-2914
Postfach 100131 Fax: +49-521-106-6411
D-33501 BIELEFELD
Germany
---------------------------(end of broadcast)---------------------------
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
