On Thu, 2007-04-26 at 22:00 -0700, Ariz Jacinto wrote:
> how's the service being used in the first place?
well, it archives detailed transaction data from KFC stores.
All sales (and other transactional) data load into the
big database (500G or so is around 2.5 years worth of data).
Aggregate tables are derived form the raw tables for faster
querying.
> i haven't dealt with DB+ReiserFS+LVM combo before
> but it's more like of a DB+JFS-LVM. yes, i removed
> the LVM layer due to a performance hit on the production
> server. we're now doing our own "snapshots" from a
> separate slave machine.
No instability there? My instability came when I was
doing very high IO operations (maintenance) on certain
tables (initially, vacuum, and later cluster).
> if you're using CentOS and you're not dependent on
> the EL features, you might as well use the vanilla
> kernel and keep yourself updated with regards to the
> filesystem (and driver) fixes (bit me once).
I post just out of curiosity :-). We're not going to
be using CentOS anymore, the server has been reinstalled
as freebsd.
> and if you're going to ask me, i might as well remove
> the RAID layer. sounds insane to some, but i really
> hate the performance hit when it is rebuilding in the
> background (heck, others can deal w/o using any
> FS journalling, hehe).
There's that. That's really a large part of our downtime,
is raid rebuild (and raid scanning of the drives to see
if there were bad blocks, didn't find any though).
I like RAID because of the redundancy, and so that the
load can be spread over all disks. for postgresql, I
realize that I can use tablespaces so I don't actually
need the large array, but I don't know if postgres will
use tablespaces at random instead of going from one
tablespace to another as tablespaces get filled up.
If it uses tablespaces at random, I wouldn't mind (since
load would be spread) but if not, then the database
would hammer the first few drives a lot as the database
grew slowly from one disk to the next disk, etc :-).
It's not my box though, and sysad prefers Raid (so do
I, for that matter), so I didn't try to drop the RAID.
> the machine maybe up in no
> time but if it can't meet the performance demand of
> the users, _it_is_just_as_good_when_it_is_offline_.
> and besides, there are lots of ways to recover from
> a failure, but that all depends on the infra.
yeah. We plan to have slony-1 replication. That way,
no downtime, just upgrade one slave to master if the
master server dies. We don't have the slave drive yet.
We'll get it sometime... :-)
tiger
--
Gerald Timothy Quimpo [EMAIL PROTECTED]
Business Systems Development, KFC/Mr Donut/Ramcar
It's a climatic shock, so we have to keep them warm with alcohol.
Yesterday we gave a bucket of vodka to one of the elephants, and
after drinking it he tore off a central heating radiator.
-- Russian zookeeper, 2006
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
[email protected] (#PLUG @ irc.free.net.ph)
Read the Guidelines: http://linux.org.ph/lists
Searchable Archives: http://archives.free.net.ph
