I posted over the weekend about this issue under the subject of "VACUUM"  but the replies I got did not help.  We have several large tables which seem to be taking 4 times as much disk space as they should?
 
Let me give a little background:
 
Every night, we do a pg_dump of our production DB and a pg_restore of this dumb into a standby DB.  What I notice is the size of the standby DB is MORE THAN HALF the size of production. 
 
For example, here is the oid2name dump of a table called users on PRODUCTION:
$ oid2name -d EBPRD1 -t users
Oid of table users from database "EBPRD1":
_______________________________
17260  = users
$ ls -l 17260*
-rw-------   1 postgres postgres 1073741824 Dec 18 17:48 17260
-rw-------   1 postgres postgres 1073741824 Dec 18 17:48 17260.1
-rw-------   1 postgres postgres 1073741824 Dec 18 17:48 17260.2
-rw-------   1 postgres postgres 165445632 Dec 18 17:51 17260.3
 
but the same table on the standby DB yields this:
 
$ oid2name -d EBPRDS1 -t users
Oid of table users from database "EBPRDS1":
_______________________________
3828262123 = users
$ ls -l 3828262123
-rw-------   1 postgres postgres 1073741824 Dec 18 16:55 3828262123
I tried a full VACUUM of the users table on production, but I didn't get any disk space back.  I also looked thru all oids and there are no pg_toast files of significance.    I am on postgres 7.2   How do I reclaim this disk space?
 
Any help would be GREATLY Appreciated!
 
thanks in advance

Reply via email to