Debian Default Moves To MariaDB

“This package build depends on libmysqlclient-dev. It should instead build-depend on default-libmysqlclient-dev metapackage, and end up having the run-time dependency of the libmysqlclient implementation Debian has chosen to use, currently MariaDB instead of Oracle MySQL.”
 
See #845905 – ruby-dataobjects-mysql: switch to build depend on the metapackage default-libmysqlclient-dev
It’s nice to see Debian on the home-stretch building the next release of GNU/Linux. They are not only providing MariaDB but also making it the default for most packages’ dependencies. I like that, freedom from tyrants is a good thing.

About Robert Pogson

I am a retired teacher in Canada. I taught in the subject areas where I have worked for almost forty years: maths, physics, chemistry and computers. I love hunting, fishing, picking berries and mushrooms, too.
This entry was posted in technology and tagged , , , , , , , . Bookmark the permalink.

10 Responses to Debian Default Moves To MariaDB

  1. oiaohm says:

    Peter, you just repeated what I mentioned, unfortunately everyone knows that you are still a dumbass.
    Dougman when you pay for maradb support they provide a set of tools.

    http://www.infoworld.com/article/3109213/open-source-tools/open-source-uproar-as-mariadb-goes-commercial.html
    There nice monitoring software that automatically take care of the Maradb databases is not free. So your choice is employ a full time DBA to do the stuff manually or pay a Maradb support contract and get the tools or attempt to do it yourself.

    Have you ever paid for Maradb dougman I would say no so you had no clue if a DBA was require or reason a DBA was required was because you had not paid for the required tools.

    Classical dougman is about time you shut up on these topics you are not current enough 2 years ago you were correct thing have changed. I said paid for support as that changed things quite a bit these days.

  2. kurkosdr wrote, “How many race conditions did the NT kernel had recently?”

    Try Google. I found this one:“An exemplary output entry yielded during the first testing iteration, which exhibited an actual vulnerability in the win32k!SfnINOUTNCCALCSIZE routine fixed by Microsoft in February 2013 is shown below:”

    Those guys wrote some software to help find race conditions and they did in NT.

    So, there you are. They don’t say how many years that vulnerability existed but it could have been since the 1990s like so many others.

    Here’s another from 2016.

    “The reason behind this is simple; Process Explorer only shows the privileges that are marked in the field Present (TokenObject->Privileges.Present). However, we can still perform privileged tasks because if the privilege is marked on the Enabled field( TokenObject->Privileges.Enabled ), Windows doesn’t perform any checks to see if it is also marked on the Present field.”

    Yes. Good one. M$ is not perfect. Far from it.

  3. Kurkosdr says:

    So, quit being alarmist. The kernel is not perfect but it is wonderful compared to alternatives.

    O really? How many race conditions did the NT kernel had recently?

    The Linux kernel community may be aware of the issues, but MS did actually put in the effort to have a fully thread safe kernel while you FOSSie guys were whining about too many locks and thread syncho in the kernel of NT…

  4. kurkosdr wrote, “Year is 2016 and teh linux kernal communitah just discovered that race conditions in kernel code is kind of a serious issue and have started addressing them. Only several decades late.”

    Nope. The kernel’s boys and girls have been aware of the dangers of race conditions for ages. See 2000

    “a race does exist in establish_pte(), when the flush_tlb happens _before_ the set_pte(), another thread might drag in the old translation on a different cpu.”

    So, quit being alarmist. The kernel is not perfect but it is wonderful compared to alternatives.

  5. kurkosdr says:

    Meanwhile in Linuxland:

    http://thehackernews.com/2016/12/linux-kernel-local-root-exploit.html?m=1

    Typical kernel contributor: “We don’t need too many locks and thread syncho, Windows has those and it’s slower”

    Year is 2016 and teh linux kernal communitah just discovered that race conditions in kernel code is kind of a serious issue and have started addressing them. Only several decades late.

  6. dougman says:

    “MariaDB, may be free, but maintaining the a large dataset is not. One will always need a paid DBA to keep the preservation and correct structure proper.” – “You can pay for support for MariaDB.”

    DERP.

    Peter, you just repeated what I mentioned, unfortunately everyone knows that you are still a dumbass.

  7. oiaohm says:

    MariaDB, may be free, but maintaining the a large dataset is not. One will always need a paid DBA to keep the preservation and correct structure proper.
    You can pay for support for MariaDB.

    Postgresql used in hospitals a lot because it does preservation so well with its WAL model and tools like barman.
    https://en.wikipedia.org/wiki/List_of_open-source_health_software

    Its interesting to follow some of those health project and find out how many cases are not deployed with a paid DBA. Postgresql deployments is way more common not to have a decanted DBA its the quality of tools for maintaining and restoring the backups means a general web site administrator can do the job most of the time and when he is in trouble your software developer gets called across.

    So if you have a decanted DBA is normally a sign of trouble in the support tools around the database you are using. This also normally means DBA is having to do lot of manual stuff to keep everything ticking over what increase risk of human error causing data loss. So must have paid DBA to use X database really should start ringing alarm bells over what is wrong with the databases tools and what are those weakness forcing to be performed manually at how much risk to your data. Must have a developer for disaster recovery/feature development and system administrators for monitoring that everything ticking is normal.

  8. dougman says:

    Coincidentally, I tried to respond earlier, but there was a database error in this site.

    MariaDB, may be free, but maintaining the a large dataset is not. One will always need a paid DBA to keep the preservation and correct structure proper.

    Well, unless you are like Robert, and are a cheapskate.

  9. Agent_Smith wrote, “Monty did a mess with Maria DB’s license”.

    Debian seems to think it’s GPLv2…:“According to the file “COPYING” all parts of this package are licenced under the terms of the GNU GPL Version 2 of which a copy is available in /usr/share/common-licenses.”

    According to Mariadb:“MariaDB is available under the terms of the GPL v2 license, just like MySQL.
     
    Some parts of the code use the BSD license or are public domain. See the header files for details.
     
    There are also a MariaDB Client Library for C and a MariaDB Client Library for Java Applications, both released with the GNU LGPL 2.1 license. Also, an old LGPL client library for MySQL 3.23 exists.”

    I don’t see a mess.

  10. Agent_Smith says:

    Monty did a mess with Maria DB’s license, so, you jump from the pan into the fire, IMO.
    Oracle is bad, but Monty is not helping either.
    Too bad, time to fork Maria SQL into something else.

    Regards,

Leave a Reply

Your email address will not be published. Required fields are marked *