I've read that MariaDB 5.2 will support virtual columns and I've read the
virtual columns' specifications:
http://forge.mysql.com/wiki/MySQL_virtual_columns_ref_manual
I think there is a problem in that document. In CREATE TABLE and in ALTER TABLE
commands you can create a virtual column. The
Hello. I have been busy, but I translated some pages and published them on a
new
wiki:
http://mariadb.pbworks.com/
The home page says that
* the wiki collects some italian unofficial translations from the askmonty.org
kb;
* anyone who wants collaborate with the translations can email me;
*
Hello
I've read in the documentation that any legal SQL expression SHOULD be allowed
in the Virtual Columns definition. I'm trying to use a (deterministic) stored
function, but I get an error:
ERROR 1642 (HY000): Function or expression is not allowed for column 'vc1'
Here's my create table:
I'd like to test the differences between InnoDB and XtraDB, but I don't want to
have 2 MariaDB instances. Is there some way to compile MariaDB with both Innodb
and XtraDB?
Federico
___
Mailing list: https://launchpad.net/~maria-discuss
Post to :
GPL 2.0 section 3 says: (read C)
3. You may copy and distribute the Program (or a work based on it,
under Section 2) in object code or executable form under the terms of
Sections 1 and 2 above provided that you also do one of the following: a)
Accompany it with the complete corresponding
5.1 documentation.
I hope that this feedback is useful.
Best greetings,
Federico Razzoli
___
Mailing list: https://launchpad.net/~maria-discuss
Post to : maria-discuss@lists.launchpad.net
Unsubscribe : https://launchpad.net/~maria-discuss
More help
(but I can't see the link
on their site) but it's not what I am looking for.
Did Oracle removed it?
Federico Razzoli
___
Mailing list: https://launchpad.net/~maria-discuss
Post to : maria-discuss@lists.launchpad.net
Unsubscribe : https
It's totally fine to report upstream bugs in MariaDB JIRA,
please don't
take offense at the tags, we just need them for our
workflow.
I don't :)
Just wanted to be sure that reporting such bugs in jira is correct.
Thanks al the team for being so responsive after a bug report!
Federico
practice.
The syntax I'm using could be wrong, but I've tried all the alternatives which
seemed to me possible, and they didn't work: DEFAULT, DEFAULT(`created_by`),
DEFAULT('created_by').
However I may still miss something, so I wanted to ask to the list before
reporting this in JIRA.
Federico
--- Ven 18/1/13, Michael Widenius mo...@askmonty.org ha scritto:
Da: Michael Widenius mo...@askmonty.org
Oggetto: re: [Maria-discuss] Reading a field's default value from a trigger
A: Federico Razzoli federico_...@yahoo.it
Cc: Maria Discuss maria-discuss@lists.launchpad.net
Data: Venerdì 18
I'm not sure that would be useful (particularly if your transactions always
leave the database in a consinstent state... which should always be the case,
unless you use non-transactional engines, but then data integrity is not
supposed to be vital).
However, there is a little trick I use when
well this may work, and may not work, you
don't set a lock between information_schema.PROCESS_LIST
and KILL command, the query can change between the time of
process_list return and the kill command, and yes, i have a
very high lucky and i killed a query that was not the right
query because
For the record, I found this patch:
https://bugs.launchpad.net/percona-server/+bug/725436
I'm not asking for it, but probably people who wants the feature proposed by
Roberto may also want this patch applied? It solves a similar problem.
Bye
Federico
Ciao,
I wanted to add some info in the CREATE PROCEDURE page in the KB, but I think
that some info about CREATE PROCEDURE clauses miss both in the kb and in the
MySQL manual.
* [NOT] DETERMINISTIC. From MySQL docs:
A routine is considered “deterministic” if it always produces the same result
Razzoli federico_...@yahoo.it
Cc: Mailing-List mariadb maria-discuss@lists.launchpad.net
Data: Lunedì 2 settembre 2013, 15:23
Hi Federico!
2013/9/2 Federico Razzoli federico_...@yahoo.it:
Ciao,
I wanted to add some info in the CREATE PROCEDURE page
in the KB, but I think that some
The documentation tells your can only use C. However:
* Antony Curtis is working on Perl Stored Procedures:
http://www.slideshare.net/AntonyTCurtis/using-perl-stored-procedures-for-mariadb
Last slide says Contributing MariaDB 10 soon. I really hope it will happen
* here there is a UDF which
A: Federico Razzoli federico_...@yahoo.it
Cc: Mailing-List mariadb maria-discuss@lists.launchpad.net
Data: Martedì 3 settembre 2013, 09:13
Hi, Federico!
On Sep 02, Federico Razzoli wrote:
Ciao,
I wanted to add some info in the CREATE PROCEDURE page
in the KB, but
I think that some
http://www.mysqlperformanceblog.com/2012/12/18/migrating-several-single-standalone-mysql-server-to-one-percona-xtradb-cluster-mariadb-to-the-rescue/
Federico
Gio 19/9/13, Roberto Spadim robe...@spadim.com.br ha scritto:
Oggetto: [Maria-discuss]
robe...@spadim.com.br ha scritto:
Oggetto: Re: [Maria-discuss] R: 'Converting' a myisam to multi-master
A: Federico Razzoli federico_...@yahoo.it
Cc: Maria Discuss maria-discuss@lists.launchpad.net
Data: Giovedì 19 settembre 2013, 20:44
Hi Federico!I was
considering a multi master (in same
:
Oggetto: Re: [Maria-discuss] R: 'Converting' a myisam to multi-master
A: Federico Razzoli federico_...@yahoo.it
Cc: Maria Discuss maria-discuss@lists.launchpad.net
Data: Venerdì 20 settembre 2013, 01:52
Hi Federico!
yes, but they don't recommend because the table lock
level (not row
I don't have benchmarks, but I think I can safely say that a bulk insert on a
InnoDB table (1 thread) is very slow on MySQL 5.7...
Federico
Dom 22/9/13, Colin Charles b...@mariadb.com ha scritto:
Oggetto: [Maria-discuss] MySQL 5.7 benchmark
A:
I found a site about MariaDB certifications:
http://www.mariadbcertification.org/
However, it looks like a sort of work in progress. Also, I don't understand if
it is something official, or at least recommended by the Foundation. And, if it
is not, is any official certification planned?
Ciao
be great, but will companies care
about them? I'm not sure, because they are not designed/approved by the MariaDB
project's people.
Federico
Il Mercoledì 23 Ottobre 2013 18:25, Colin Charles b...@mariadb.com ha scritto:
Hi!
On 24 Oct 2013, at 00:11, Federico Razzoli federico_...@yahoo.it wrote
Did you run SET ROLE? Roles are not activated automatically. Users must declare
they want to use a particular role.
Federico
Il Mercoledì 13 Novembre 2013 18:37, Adam Scott adam.c.sc...@gmail.com ha
scritto:
Roles are supported according to
Hi! MySQL 5.6 returns a warning for wrong datatypes, with any SQL_MODE:
mysql CREATE TABLE t (c TIMESTAMP) ENGINE=InnoDB;
Query OK, 0 rows affected (0,31 sec)
mysql SELECT c FROM t WHERE c = 1;
Empty set, 1 warning (0,00 sec)
Warning (Code 1292): Incorrect datetime value: '1' for column 'c' at
I found contradictory information about whom the MariaDB trademark belongs to:
The MariaDB Foundation also holds the trademark of the MariaDB server and owns
mariadb.org. This ensures that the official MariaDB development tree will
always be open for the MariaDB developer community.
IMO, MariaDB (like MySQL) would better issue a warning when trying to create a
hash index and the engine does not support it. I'm not surprised if someone
thinks that all engines support hash :)
Regards
Federico
Gio 6/3/14, Sergei Golubchik
I agree with all Paul Larsen's thoughts about stored programs.
@Roberto, my personal opinion:
If you you are counting the occurrences of ONE char, you can use this
expression:
LENGTH(col_name) - LENGTH(REPLACE(col_name, 'a', ''))
C functions, even when used in a stupid way (like in this
When a PHP function does something you cannot do in SQL, you could call PHP
from your queries using the Gearman UDF.
https://launchpad.net/gearman-mysql-udf
In most cases this should be faster than a stored function. I hope that in the
future this will change, but who knows...
Regards,
Carsten, your example is very interesting. I didn't expect that 10 native funcs
calls can be 242 times faster (in a real/realistic use case)!
A suggestion: one can use macros. strip_digits() can become a macro, so the
code will be readable and there won't be performance problems. I use GNU m4
Hello list
I found out that EXPLAIN SELECT aggregate_func() FROM connect_table returns a
rows value which is much greater than the rows in the table. For example:
MariaDB [test] SELECT COUNT(*) FROM t;
+--+
| COUNT(*) |
+--+
| 100 |
+--+
1 row in set (0.49 sec)
and CONNECT
A: Federico Razzoli federico_...@yahoo.it
Cc: maria-discuss@lists.launchpad.net
Data: Mercoledì 26 marzo 2014, 20:44
Hi, Federico!
On Mar 23, Federico Razzoli
wrote:
Hello list
I found out that EXPLAIN SELECT
aggregate_func() FROM connect_table returns a
rows value
Hi list
This is just a curiosity, but anyway:
Of course I understand the need for the FLUSH ... FOR EXPORT syntax for InnoDB.
But why does MyISAM support it? In which way is it better than WITH READ LOCK?
Regards,
Federico
___
Mailing list:
: Re: [Maria-discuss] EXPLAIN and CONNECT
Para: Federico Razzoli federico_...@yahoo.it
CC: maria-discuss@lists.launchpad.net
Fecha: lunes, 12 de mayo, 2014 15:40
Hi, Federico!
On Mar 27, Federico Razzoli
wrote:
Hi Sergei
You can repeat this very easily:
MariaDB [test]
CREATE
If statements like CREATE ROLE were allowed in prepared statements, this could
easily be done with a stored procedure.
Regards
Federico
El jue, 22/5/14, Sergei Golubchik s...@mariadb.org escribió:
Asunto: Re: [Maria-discuss] Role template
Para:
How could Spider invalidate the query cache when necessary? The remote servers
don't know about your Spider table. Consider using the query cache on the
remote server, if it's really necessary.
Federico
El jue, 29/5/14, Roberto Spadim
I don't use it in production, but this may change in the future. The problem is
that I cannot enable it when I need it, if I don't want to restart the server.
Regards
Federico
El mar, 3/6/14, Colin Charles co...@mariadb.org escribió:
Asunto:
Duplicate data cannot be inserted in MariaDB, in no cases.
PostgreSQL has a different behaviour: transactions can optionally be
deferred, which means that integrity checks will be done on commit. This
means that you can temporarly insert inconsistent data, and fix them before
commit.The same
Pantelis Theodosiou yperc...@gmail.com:
On Thu, Jun 12, 2014 at 9:57 AM,
Federico Razzoli federico_...@yahoo.it
wrote:
Duplicate data cannot be
inserted in MariaDB, in no cases.
PostgreSQL has a different behaviour: transactions can
optionally be deferred, which means
It's the easiest and least locking way, I think. But if you want more detailed
info:
SELECT * FROM information_schema.PROCESSLIST
Regards
Federico
El dom, 15/6/14, Tanstaafl tansta...@libertytrek.org escribió:
Asunto: Re: [Maria-discuss] Easy way
That page is interesting! Can I suggest a couple more information to show? I
see that you collect stats about fulltext and GIS - if possible, it would be
interesting to know which storage engines people use for these features.
Regards
Federico
El
I know nothing about SOGo (this is a MariaDB list), but I searched the proper
mailing list for you, and it seems that someone already had your problem:
http://www.mail-archive.com/users%40sogo.nu/msg18460.html
Regards,
Federico
Mer 9/7/14,
Hi Roberto
1) Can't you use a relational table? (obvious suggestion, so probably the
answer is no. but since MariaDB is relational, this is the main source of your
troubles...)
2) If you need 1 row, why should an application send a query like that? It
should send the simple SELECT first, and
In this MySQL manual page:
http://dev.mysql.com/doc/refman/5.6/en/implicit-commit.html
I read:
UNLOCK TABLES commits a transaction only if any tables currently have been
locked with LOCK TABLES to acquire nontransactional table locks.
However, this doesn't seem to be the case, at least with
] transactions and UNLOCK TABLES
A: Federico Razzoli federico_...@yahoo.it
Cc: Maria Discuss maria-discuss@lists.launchpad.net
Data: Lunedì 22 settembre 2014, 22:10
good
question, at least here the implicity commit is with LOCK
table and not unlock table
CREATE
OR REPLACE TABLE m (c INT
robe...@spadim.com.br ha scritto:
Oggetto: Re: [Maria-discuss] transactions and UNLOCK TABLES
A: Federico Razzoli federico_...@yahoo.it
Cc: Maria Discuss maria-discuss@lists.launchpad.net, maria-docs
maria-d...@lists.launchpad.net
Data: Martedì 23 settembre 2014, 00:14
better
explained
] transactions and UNLOCK TABLES
A: Federico Razzoli federico_...@yahoo.it
Cc: Maria Discuss maria-discuss@lists.launchpad.net
Data: Martedì 23 settembre 2014, 01:46
well from
kb what i understand is:to use unlock with innodb (the
engine you use with both tables), you should
SET innodb_table_locks=1
Let's make the snippet simpler. Only one table, non-transactional, please look
at the difference from what docs say and the real behaviour. If one could
explain what UNLOCK TABLES exactly does, it would be great.
MariaDB [test] SELECT @@in_transaction, @@autocommit;
Mar 23/9/14, Elena Stepanova ele...@montyprogram.com ha scritto:
Oggetto: Re: [Maria-discuss] R: transactions and UNLOCK TABLES
A: Federico Razzoli federico_...@yahoo.it,
maria-discuss@lists.launchpad.net
Data: Martedì 23 settembre 2014, 11:51
Hi Federico,
On 23.09.2014
Hello.
I've read in this page
https://mariadb.com/kb/en/mariadb/documentation/sql-commands/table-commands/optimize-table/
that OPTIMIZE TABLE defragments InnoDB tables in 10.1. Good news! Just some
questions: What locks are acquired? Is the table copied? Is the operation
atomic?
Regards
Not tested, but should work:
1) use only 1 binary and only 1 instance
2) make a startup script. when starting in test mode, it will copy your datadir
to a ram disk, and will create/overwrite a symlink to the ramdisk. when in
normal mode, the script will copy from ram to disk and the symlink will
Please, people, note that this fight has flooded a lot of people's mailboxes
with non-technical or out-of-topic mails. If the thread continues, this could
be a problem for some of us.
Regards
Federico
___
Mailing list:
Hi all
Would someone kindly explain (or document) how @@thread_cache_size affects the
pool-of-threads? Do values 0 make sense?
Regards
Federico
___
Mailing list: https://launchpad.net/~maria-discuss
Post to : maria-discuss@lists.launchpad.net
Sorry, I missed that part, as well as previous Jean's question.
Thank you
Federico
Mer 3/12/14, vvaintr...@gmail.com vvaintr...@gmail.com ha scritto:
Oggetto: Re: [Maria-discuss] thread_cache_size and buffer pool
A: Federico Razzoli federico_
Hi.
In MySQL 5.7 documentation:
https://dev.mysql.com/doc/refman/5.7/en/innodb-parameters.html
I read:
As of MySQL 5.7.5, the InnoDB storage engine can no longer be disabled, and
the --innodb=OFF and --skip-innodb options are deprecated and have no effect.
Their use results in a warning.
Roberto,
The meaning of an SQL is not engine-dependent. But here you have a timing
problem.
As Elena explained, Aria allows concurrent inserts to the same table. As a
consequence, if MAX(id) is 100, several threads could try to insert 101,
resulting in duplicate key errors.
Your application
:
Oggetto: Re: [Maria-discuss] Doubt about 'atomic' insert
A: Federico Razzoli federico_...@yahoo.it
Cc: Maria Discuss maria-discuss@lists.launchpad.net
Data: Lunedì 15 dicembre 2014, 18:23
Hi
guys,
i agree with you federico, sql is a
language, only, and engine do what it's supossed to
do
Hi!
Galera claims that it only supports InnoDB tables (and MyISAM, experimentally).
What about storage engines that add functionalities upon another table? For
example, can I have a Mroonga or OQGRAPH or CONNECT/PROXY table built over an
InnoDB table?
As far as I understand this should be
Dom 28/12/14, Reindl Harald h.rei...@thelounge.net ha scritto:
Oggetto: Re: [Maria-discuss] R: 5.5.x vs 10.x
A: maria-discuss@lists.launchpad.net
Data: Domenica 28 dicembre 2014, 19:00
Am
28.12.2014 um 18:17 schrieb Federico Razzoli:
10.0 is the newest stable
Wikipedia's explanation sounds like a way to say nothing, while still
pretending they are saying something. An information does not exist in the
database? This is obvious for anyone! I don't know a single person stupid
enough to think that NULL is a valid name or telephone number :) The open
Mar 3/2/15, Roberto Spadim robe...@spadim.com.br ha scritto:
Oggetto: Re: [Maria-discuss] Restated: JSON cannot represent binary data
A: Federico Razzoli federico_...@yahoo.it
Cc: Justin Swanhart greenl...@gmail.com, Tom Worster f...@thefsb.org,
maria
Hi
I agree that EAV is a pain in the ass. Having several very similar tables that
represent flavors of the same parent entity is a pain too. Dynamic Columns
are logically simple.
I think that the problem is that people don't really know them. And if one
tries to explore them, he finds a
I don't want to show critisism against projects like this - I hope that this is
clear. There are many debuggers, and they are more than welcome. Even Oracle
has a debugger, if you use Windows and Visual Studio (sigh).
But I still think that MariaDB needs a native debug API, which fully supports
This JIRA task doesn't mention GSoC, but it seems to be what you are asking for:
https://mariadb.atlassian.net/browse/MDEV-5199
Hope it helps.
Federico
Mar 10/3/15, Delveri chick chickdelv...@gmail.com ha scritto:
Oggetto: [Maria-discuss] GSOC
I made some quick test to show what I mean by performance problem. Note that
I'm not saying the the first and the second test cases should perform equally.
But there is too much difference - see the conclusion.
1)
MariaDB [test] SELECT BENCHMARK(5000, (SELECT 1));
...@webyog.com ha scritto:
Oggetto: Re: [Maria-discuss] stored programs
A: Federico Razzoli federico_...@yahoo.it
Cc: Maria Discuss maria-discuss@lists.launchpad.net
Data: Martedì 3 marzo 2015, 12:48
1) +1 for
ability to
prepare a statement from a local variable what I
requested 6½ years
/3/15, Justin Swanhart greenl...@gmail.com ha scritto:
Oggetto: Re: [Maria-discuss] stored programs
A: Peter Laursen peter_laur...@webyog.com
Cc: Federico Razzoli federico_...@yahoo.it, Maria Discuss
maria-discuss@lists.launchpad.net
Data: Martedì 3 marzo 2015, 14:28
You
probably have
] stored programs
A: Peter Laursen peter_laur...@webyog.com
Cc: Federico Razzoli federico_...@yahoo.it, Maria Discuss
maria-discuss@lists.launchpad.net
Data: Martedì 3 marzo 2015, 15:04
Hi,
The server
parses the function into a form of bytecode which is then
cached until the routine
, Federico!
On Mar 03, Federico Razzoli wrote:
Reading 10.0.3 release notes:
https://mariadb.com/kb/en/mariadb/mariadb-1013-release-notes/
I see that IF EXISTS, IF NOT EXISTS and OR REPLACE
are now almost
consistent. Almost means that... OR
REPLACE still doesn't apply
Mar 3/3/15, Sergei Golubchik s...@mariadb.org ha scritto:
Oggetto: Re: [Maria-discuss] stored programs
A: Federico Razzoli federico_...@yahoo.it
Cc: maria-discuss@lists.launchpad.net
Data: Martedì 3 marzo 2015, 10:48
Hi, Federico!
On Mar 03, Federico Razzoli
wrote
/15, Reindl Harald h.rei...@thelounge.net ha scritto:
Oggetto: Re: [Maria-discuss] doubt about index
A: maria-discuss@lists.launchpad.net
Data: Martedì 23 giugno 2015, 01:31
Am 23.06.2015 um 01:20 schrieb Federico
Razzoli:
I still don't understand
the matter.
teste is a unique
I still don't understand the matter.
teste is a unique index, because part of it is unique. But it is not declared
as unique, and MariaDB is not suppose to read our minds.
The query written by Reindl looks realistic, but it would use the primary key,
and I would be disappointed if it used
That's a strange index. In InnoDB, tables are organized by primary key, so ALL
indexes contain primary keys values.
In other storage engines... well, I can't think a use case when that index
would make sense. Maybe a more realistic example would help :)
Federico
Ven 12/6/15, Rhys Campbell rhys.campb...@tradingscreen.com ha scritto:
Oggetto: RE: [Maria-discuss] ON DELETE/UPDATE/INSERT Triggers
A: Federico Razzoli federico_...@yahoo.it, Geoff Montee
geoff.mon...@gmail.com, Roberto Spadim robe...@spadim.com.br
Cc: MariaDB discuss maria
Hi
TABLE_TYPE = BIN or FIX.
READONLY=1
I think that you can skip the header with HEADER=1. But I'm not sure that this
will work with these particular table types.
Sab 13/6/15, Roberto Spadim robe...@spadim.com.br ha scritto:
Oggetto:
In the meanwhile, we can use ON DELETE triggers. But the request is a good
idea, because it is a common operation when we want to move some rows to
history tables.
Federico
Gio 11/6/15, Roberto Spadim robe...@spadim.com.br ha scritto:
Oggetto:
I'm not sure. Flexviews should be a good solution.
Or you can used triggers to replicate the table instantly.
I don't understand your idea with SPIDER, how will it help you?
Regards
Federico
Ven 3/7/15, Roberto Spadim robe...@spadim.com.br ha
Hi all
I remember these slides and this worklog about supporting external languages
for stored procedures:
* http://www.slideshare.net/AntonyTCurtis/external-stored-proceduresformysql2009
*
@lists.launchpad.net]
On Behalf Of Federico Razzoli
Sent:
Wednesday, July 22, 2015 10:54 PM
To: maria-discuss@lists.launchpad.net
Subject: [Maria-discuss] stored procedures
external languages
Hi
all
I remember these slides
and this worklog about supporting external languages
Very interesting, thank you!
However, the link #8 seems to be wrong (it leads to "MaxScale", not to "PCRE").
Federico
Sab 24/10/15, Colin Charles ha scritto:
Oggetto: [Maria-discuss] MySQL Meetup NL slides
A: "Maria Discuss
If you are talking about donations to the foundation (not the company) I agree
that it's a great idea. If small donations are welcome, many people will
probably donate (and I'm one of them).
Just a hint: maybe you could even advertise merchandising in the download page.
Regards
Federico
Standard SQL has triggers FOR EACH STATEMENT, but MariaDB has not. I would use
a stored procedure instead.
Regards
Federico
Ven 13/11/15, Roberto Spadim ha scritto:
Oggetto: [Maria-discuss] doubt trigger
A: "Maria Discuss"
Just a personal curiosity: if one wants to add a feature request about TokuDB
and a feature that is only supported in MariaDB (not MySQL or Percona), where
should the feature request be added?
Federico
Mar 3/11/15, Roberto Spadim
Personal opinion:
I know that nowdays we're all subscribed to multiple Google services, but this
shouldn't be required to be part of a list. I'd prefer a solution other than
Google Groups.
Regards
Federico
Mer 14/10/15, Colin Charles
And anyway, that's how the real world works... no matter if one understands the
reason or not :)
Federico
Dom 12/7/15, Justin Swanhart greenl...@gmail.com ha scritto:
Oggetto: Re: [Maria-discuss] ssd and hdd
A:
Cc:
A new version of CONNECT package was released, but you have still 10.1.9. The
packages don't conflict, but versions are incompatible.
Just run apt-get upgrade and CONNECT will work.
Regards
Federico
Gio 24/12/15, Karol Krasňan ha
InnoDB is different from MyISAM. It crashed intentionally, as stated in the log
you posted, to avoid data corruption propagation.
Take a look at this page, so next time you can try to repair tables:
http://dev.mysql.com/doc/refman/5.7/en/forcing-innodb-recovery.html
Regards,
Federico
Weird issue. I'm trying to figure out how the connection pool can add an
overhead to a single query. The only case I can think is if you use a network
storage engine (Federated, Connect, Spider) pointing to tables on localhost. Is
this the case?
More likely... are you sure this problem is
Hi Justin!
The name reminds me a a fatal trap for developers and sys admins (async npm
module), but the tools sounds useful.
Is it publicly available? Is it entirely implemented via stored procedures +
event scheduler?
bye
Federico
Mer 24/2/16,
Hi!
I see that MariaDB switched to the new JIRA. However this redirect has not been
updated:
http://mariadb.org/jira
Regards
Federico
___
Mailing list: https://launchpad.net/~maria-discuss
Post to : maria-discuss@lists.launchpad.net
Unsubscribe
I never had the login problem described by Justin, so I tested the GitHub login
just for curiosity. What happened seems to me weird:
* I logged out
* I logged in via GitHub for the first time. An email verification is required
(why?)
* I didn't verify the mail
* I logged in with my email &
While compiling MariaDB on FreeBSD, I've noted that it can optionally be
compiled with spinlocks instead of mutexes. What's the best choice? Any
experience?
Regards
Federico
___
Mailing list: https://launchpad.net/~maria-discuss
Post to :
林澤宇, there is clearly some difference between your servers.
Same version?
Same storage engine (InnoDB)?
Same indexes?
Memory is full, maybe swapping?
If all these things are ok, check your settings (SHOW VARIABLES).
Federico
Gio 10/3/16, 林澤宇
That could be a problem if DNS is not reliable, but I never heard of a 10
seconds DNS request.
I'd strongly recommend to assign more memory to the buffer pool. If InnoDB is
the main storage engine you use, 1.5G should be ok.
Swap is evil for a database, so you should make sure that loading a
Stored procedures are slow in MariaDB and MySQL. I reported a bug for this, and
Elena Stepanova's comment explains the reason, or maybe one of the reasons. The
bug is:
https://jira.mariadb.org/browse/MDEV-8254
MariaDB 10.1 is not slower than other versions. From my tests, MySQL 5.7 is
almost
Hi,
In current MariaDB versions, InnoDB has 2 types of tablespaces:
* system tablespace
* per-table tablespace
There is a variable called innodb_file_per_table, you can change it at runtime.
When a new table is created with innodb_file_per_table=1, the table has its own
tablespace (an .ibd
until this issue is
sorted.
Kind Regards,
Bruce Carlson
-Original Message-----
From: Federico Razzoli [mailto:federico_...@yahoo.it]
Sent: Tuesday, 19 April 2016 10:41 AM
To: maria-discuss@lists.launchpad.net; Bruce Carlson
Subject: R: [Maria-discuss] Mariadb 10.
Hi all
I see that 10.2 still has the innodb_fake_changes variable (OFF by default).
That feature was stable in Percona Server 5.5, is beta in 5.6, and was removed
in 5.7.
Is it still there just for compatibility with old configuration files, or is
there still a reason to use it? And in that
Well, I can't think many cases where InnoDB is not the default engine. But, if
it is not, why should you use it for temporary tables?
If you mainly use TokuDB, maybe don't want to mix the engines. If you use
MyISAM, maybe you don't want transactional temporary tables.
I still think that MEMORY
Hi Ian
I beleive that there is a difference if default_storage_engine != innodb,
because in that case, default_tmp_storage_engine will not be InnoDB.
It is much less clear to me why default_tmp_storage_engine doesn't default to
MEMORY in both MariaDB and MySQL :)
Federico
1 - 100 of 138 matches
Mail list logo