On 01/16/2015 11:22 AM, Enrico Bianchi wrote:
https://drive.google.com/file/d/0ByfjZX4TabhocUg2MFJ6a21qS2M/view?usp=sharing
Note: due an error in dump script, if you are in Linux/Unix environment,
use this command for uncompressing the file:
bzip2 -d -c comment_test.dump.bz2 |sed -e
On 01/16/2015 09:58 AM, Enrico Bianchi wrote:
I've asked permission for these data
I've obtained the permission, here is available a subset of data large
enough to replicate the problem (note: you can simply run the query
without the where clause):
https://drive.google.
On 01/16/2015 01:19 AM, John R Pierce wrote:
you didn't do EXPLAIN ANALYZE, so your query plan statistics are all
estimates.
I know, but the EXPLAIN ANALYZE has the same problem of the query
Enrico
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your s
On 01/16/2015 02:18 AM, Tom Lane wrote:
Can we see the map?
This is the log when executing the query with a subset of data:
< 2015-01-16 08:47:43.517 GMT >DEBUG: StartTransactionCommand
< 2015-01-16 08:47:43.517 GMT >DEBUG: StartTransaction
< 2015-01-16 08:47:43.517 GMT >DEBUG: name: unnamed
I have this situation:
Machine:
VPS with CentOS 6.6 x86_64
64GB of RAM
2GB of swap (unused)
Ulimit settings:
postgressoftnproc 2047
postgreshardnproc 16384
postgressoftnofile 1024
postgreshardnofile 65536
postgreshard
