Re: POSTGRES/MYSQL - Mailing list pgsql-general

From Christopher Browne
Subject Re: POSTGRES/MYSQL
Date
Msg-id CAFNqd5Xc31Psf3ecEQ8b1o99V6--aE2BX2QVida3x_nvXpNsPA@mail.gmail.com
Whole thread Raw
In response to Re: POSTGRES/MYSQL  (Benedict Holland <benedict.m.holland@gmail.com>)
Responses Re: POSTGRES/MYSQL  (Ron <ronljohnsonjr@gmail.com>)
List pgsql-general
On Tue, 12 Mar 2019 at 12:53, Benedict Holland
<benedict.m.holland@gmail.com> wrote:
> I am not saying it is not well documented. I am saying that it isn't ACID compliant, which it isn't, as they
document.

I *love* the notion of being able to roll back DDL, but it has long
been common for DDL to *not* be transactional even with some of the
Big Expensive Databases (such as the one whose name begins with an
"O").

Up until version 11.something, "Big O" apparently did NOT have this,
and MS SQL Server didn't in version 2008.

https://docs.oracle.com/cd/B19306_01/server.102/b14220/transact.htm
https://docs.microsoft.com/en-us/previous-versions/sql/sql-server-2008-r2/ms189122(v=sql.105)

Of course, those are somewhat old versions.  But nobody would have
claimed those systems not to be "ACID Compliant" at the time; you're
setting the bar a bit too high.

Someone's asking the merits of PostgreSQL versus MySQL; it certainly
*is* possible to overplay the case.

I'm perfectly happy with a claim like...

 "PostgreSQL does transactional DDL, which we find quite valuable, and
while MySQL supports ACID for data manipulation, with suitable choice
of storage engines, there is not the same capability to be able to
roll back DDL within a transaction."
-- 
When confronted by a difficult problem, solve it by reducing it to the
question, "How would the Lone Ranger handle this?"


pgsql-general by date:

Previous
From: Vijaykumar Jain
Date:
Subject: Re: [External] Re: xmin and very high number of concurrent transactions
Next
From: Ron
Date:
Subject: Re: POSTGRES/MYSQL