Em 28/10/2012 01:35, Gavin Flower escreveu:
> On 28/10/12 12:29, Leif Biberg Kristensen wrote:
>> Søndag 28. oktober 2012 01.17.45 skrev Gavin Flower :
>>> Also note that for features that are obviously complicated or advanced,
>>> Postgres tends to a lot better than MySQL.
>> It's like comparing BASIC to C. BASIC has a low threshold, but you
>> will very
>> quickly bump your head against the wall.
>>
>> MySQL, the BASIC of db engines?
>>
>> regards, Leif
>>
>>
> I have used both MySQL & Postgres: I feel a lot more comfortable with
> Postgres, as it seems to be both simpler and more sophisticated.
> MySQL has several DB engines for different purposes, Postgres has just
> one that appears to be more capable than the collection of features
> from all the MySQL DB engines combined. In the last 12 years I've
> gone looking for comparisons between them 3 times; and each time,
> Postgres comes out better overall.
>
And I must add: even using InnoDB, MySQL allows violation of relational
integrity (just put it under heavy transactional load).
I've suffered this pain in the past even with MySQL 5.1. This problem
does not happens with PostgreSQL.
The only occasion I had duplications in PostgreSQL was during a heavy
data load. At that time, I used RSync to copy data to another server
with the command:
rsync -azv /var/lib/pgsql/9.1/data root@127.0.0.1:/var/lib/pgsql/9.1/
Do you see my mistake above? Yes, this causes duplicate records (and
after some time, complete database corruption in PostgreSQL).
Is just db admin stupidity (I can tell, because I was caused by my
self). Would be nice if we can shield PostgreSQL against my personal
silliness.
Edson
>
> Cheers,
> Gavin
>
>
