Mindaugas Riauba wrote:
When a row is orphaned it's added to a list of possibly available rows.
When a new row is needed the list of possible rows is examined and the
first one with a transaction id less then the lowest running transaction
id is chosen to be the new row? These rows can be in a
Hi,
I´m trying to tune a linux box with a 12 GB database and4 GB RAM. First of all I would like to stopthe swapping, so the shared_buffers and sort_mem were decreased but even so it started swapping two hours after DBMS started up.
I would like to know some suggestions how to discover why is it
Carlos Henrique Reimer [EMAIL PROTECTED] writes:
I would like to know some suggestions how to discover why is it swapping?
Zero swap-in rate and swap-out rates in the single digits do not
constitute a swapping problem. It's reasonably likely that that
traffic isn't even coming from Postgres,
Hi,
I am currently trying to speed up the insertion of bulk loads to my database. I
have fiddled with all of the parameters that I have seen suggested(aka
checkpoint_segments, checkpoint_timeout, maintinence_work_mem, and shared
buffers) with no success. I even turned off fysnc with no effect
[Carlos Henrique Reimer - Wed at 03:25:15PM -0300]
I´m trying to tune a linux box with a 12 GB database and 4 GB RAM. First
of all I would like to stop the swapping, so the shared_buffers and sort_mem
were decreased but even so it started swapping two hours after DBMS started
up.
I would
[Tobias Brox - Wed at 09:22:17PM +0200]
I'd trust linux to handle swap/cache sensibly. Eventually, become involved
with kernel hacking ;-)
Of course, there are also some files in /proc/sys/vm that you may want to
peek into, for tuning the swapping. Particularly, at later 2.6-kernels (I'm