February 2015 - discuss - lists.mariadb.org

[Maria-discuss] R: Re: R: Re: R: Re: Switch from rsync to xtrabackup
by absolutely_free＠libero.it 18 Feb '15

18 Feb '15

Ok, thank you all, switch to xtrabackup-v2 accomplished :)

2 1

Re: [Maria-discuss] MariaDB 10.0.16 now available
by Nirbhay Choubey 18 Feb '15

18 Feb '15

Hi Raina, On Tue, Feb 17, 2015 at 1:04 PM, Raina Masand <rmasand(a)pivotal.io> wrote: > Hi Nirbhay, > > Now that Galera 25.3.9 is available, we have tried swapping it in with our > MariaDB 10.0.16 cluster. Seems to work so far, but wondering whether there > are plans to vet this Galera release as the officially recommended version? > I see that the currently recommended Galera version is 25.3.5, which was > released quite a while ago. Any news as to their compatibility would be > appreciated. > I have just finished running some tests on 10.0.16 with galera-25.3.9 with no failure. So, you should be good. By the way, we have not yet released galera-25.3.9 packages, planning to release them in upcoming MariaDB Galera cluster release. Best, Nirbhay > > Thanks, > > Raina and Lyle > Cloud Foundry Services Team > > On Tue, Feb 3, 2015 at 3:15 PM, Nirbhay Choubey <nirbhay(a)mariadb.com> > wrote: > >> Hi! >> >> On Tue, Feb 3, 2015 at 5:25 PM, Raina Masand < >> cf-core-services-eng(a)pivotal.io> wrote: >> >>> Hello, >>> >>> We are looking to use Galera 25.3.9 with our MariaDB cluster, which >>> claims to improve the stability of communication between nodes in the event >>> of network flakiness. Do you know if it will be compatible with MariaDB >>> 10.0.16? The release notes for MariaDB Galera Cluster 10.0.16 seem to >>> suggest that this should be safe, but it would be useful to know the >>> official recommendation. >>> >> >> MariaDB Galera 10.0.16 should be compatible with any 25.x.x galera >> versions. But, 25.3.9, in particular, has not been tested yet with MGC. >> >> Best, >> Nirbhay >> >> >>> Thanks, >>> >>> Raina and Lyle >>> Cloud Foundry Services Team >>> >>> On Wed, Jan 28, 2015 at 6:45 AM, Daniel Bartholomew <dbart(a)mariadb.com> >>> wrote: >>> >>>> On Wed, Jan 28, 2015 at 9:39 AM, Peter Laursen < >>>> peter_laursen(a)webyog.com> wrote: >>>> > YaST "Online Update" interface did not (both yesterday and two hours >>>> ago), >>>> > does not and will not identify that there is an upgrade available for >>>> > MariaDB 10.0.15, when there actually was/is. I cannot decide it this >>>> is an >>>> > issue with the RPM packaging or the repo configuration. But something >>>> is not >>>> > correct for SuSE. >>>> >>>> Ok, I'll look into it. >>>> >>>> Thanks! >>>> >>>> -- >>>> Daniel Bartholomew, MariaDB Release Manager >>>> MariaDB | http://mariadb.com >>>> >>>> _______________________________________________ >>>> Mailing list: https://launchpad.net/~maria-discuss >>>> Post to : maria-discuss(a)lists.launchpad.net >>>> Unsubscribe : https://launchpad.net/~maria-discuss >>>> More help : https://help.launchpad.net/ListHelp >>>> >>> >>> >>> _______________________________________________ >>> Mailing list: https://launchpad.net/~maria-discuss >>> Post to : maria-discuss(a)lists.launchpad.net >>> Unsubscribe : https://launchpad.net/~maria-discuss >>> More help : https://help.launchpad.net/ListHelp >>> >>> >> >

1 0

[Maria-discuss] R: Re: R: Re: Switch from rsync to xtrabackup
by absolutely_free＠libero.it 18 Feb '15

18 Feb '15

Hi Nirbhay, thank you very much for your answer. This cluster is already in production, with few databases (for mail user authentication, radius authentication, and so on). My question is: what would will happen if I change wsrep_sst_method on one node? How the other node can be "synced" if wsrep_sst_method is different (even for a few seconds)? Thank you, regards

2 1

[Maria-discuss] R: Re: Switch from rsync to xtrabackup
by absolutely_free＠libero.it 18 Feb '15

18 Feb '15

Hi, thank you for your reply. What is the correct order for this step? For example: 1) shutdown node 1 2) change config to xtrabackup-v2 on node 1 3) restart node 1? 4) shutdown node 2 5) change config to xtrabackup-v2 on node 2 6) restart node 2 ..and what about garbd on the third node? :) Thank you very much ----Messaggio originale---- Da: nirbhay(a)mariadb.com Data: 17/02/2015 14.22 A: "absolutely_free(a)libero.it"<absolutely_free(a)libero.it> Cc: "MariaDB discuss"<maria-discuss(a)lists.launchpad.net> Ogg: Re: [Maria-discuss] Switch from rsync to xtrabackup Hi! On Tue, Feb 17, 2015 at 6:07 AM, absolutely_free(a)libero.it <absolutely_free(a)libero.it> wrote: Hi, I am using two nodes (Debian 7.8 64bit) with MariaDB (mariadb-galera-server 5.5.41) and garbd on a third node. Now I'm using rsync method, I would like to switch to xtrabackup to avoid "lock" issue. Is it possible? Yes, the lock is taken only for syncing system tables. Should I simply install percona package and change this configuration? Right, the package name is "percona-xtrabackup". wsrep_sst_method=rsync to wsrep_sst_method=xtrabackup ? Or rather go for "xtrabackup-v2".https://mariadb.com/kb/en/mariadb/galera-cluster-system-variables/#wsrep_sst_method SST methods other than "rsync" also require "wsrep_sst_auth" variables to be set.https://mariadb.com/kb/en/mariadb/galera-cluster-system-variables/#wsre… Best,Nirbhay Thank you very much _______________________________________________ Mailing list: https://launchpad.net/~maria-discuss Post to : maria-discuss(a)lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-discuss More help : https://help.launchpad.net/ListHelp

2 1

[Maria-discuss] Network tuning
by absolutely_free＠libero.it 18 Feb '15

18 Feb '15

Hi all, according to this URL: http://www.fromdual.com/huge-amount-of-time-wait-connections is it recommended to change default behavior of tcp_tw_reuse and tcp_tw_recycle? I am using two MariaDB nodes with haproxy, keepalived and garbd on third node. Thank you very much

1 0

[Maria-discuss] INNODB_BUFFER_PAGE
by Rhys Campbell 17 Feb '15

17 Feb '15

Hi All, I've been playing with the innodb buffer pool save / load feature in MariaDB 10.0.14. I initially thought it wasn't working but I've since identified the cause of my confusion The buffer pool was being populated with the correct number of pages but no table/index name values were given in information_schema. INNODB_BUFFER_PAGE. These details are populated as soon as the database has been access. Steps to repeat. 1. Ensure buffer pool save/load is on. 2. Record the SPACE & PAGE_NUMBER details of a page in information_schema.INNODB_BUFFER_PAGE for one of your databases. 3. Stop MariaDB. 4. Start MariaDB 5. Allow time for buffer pool to load... 6. SELECT * FROM information_schema.INNODB_BUFFER_PAGE WHERE SPACE = 157 AND PAGE_NUMBER = 99212; <- NULL values for TABLE_NAME / INDEX_NAME 7. Execute USE <db_name> # Where db_name is the database containing the above page. 8. Repeat above select and TABLE_NAME and INDEX_NAME will be populated correctly. Now, on a busy database this wouldn't be an issue but it did throw me for a while on my dev box. I have found a MySQL bug report... http://lists.mysql.com/announce/847 "When running a query on INFORMATION_SCHEMA.INNODB_BUFFER_PAGE that requested table_name and index_name values, query results would include index pages without table_name or index_name values. (Bug #14529666)" But the bug has been deleted so I can't check the full details to see if this is resolved or still an issue. As an aside the logging for the buffer pool loading feature appears to use a different datetime format... 150217 18:03:37 [Note] InnoDB: Percona XtraDB (http://www.percona.com) 5.6.20-68.0 started; log sequence number 103721589973 2015-02-17 18:03:37 7f53285f8700 InnoDB: Loading buffer pool(s) from /log/file/path/... Rhys Campbell Database Administrator TradingScreen, Inc. 23 York House, 5th Floor London WC2B 6UJ Email: rhys.james.campbell(a)googlemail.com<mailto:rhys.james.campbell@googlemail.com> Follow TradingScreen on Twitter<http://twitter.com/#!/TradingScreen> , Facebook<http://www.facebook.com/pages/TradingScreen/214046251945650> and our blog Trading Smarter<tradingsmarter.tradingscreen.com> This message is intended only for the recipient(s) named above and may contain confidential information. If you are not an intended recipient, you should not review, distribute or copy this message. Please notify the sender immediately by e-mail if you have received this message in error and delete it from your system.

1 0

[Maria-discuss] Switch from rsync to xtrabackup
by absolutely_free＠libero.it 17 Feb '15

17 Feb '15

Hi, I am using two nodes (Debian 7.8 64bit) with MariaDB (mariadb-galera-server 5.5.41) and garbd on a third node. Now I'm using rsync method, I would like to switch to xtrabackup to avoid "lock" issue. Is it possible? Should I simply install percona package and change this configuration? wsrep_sst_method=rsync to wsrep_sst_method=xtrabackup ? Thank you very much

2 1

Re: [Maria-discuss] MariaDB 10 multisource replication from AWS MySQL 5.6 RDS and Percona Server 5.6
by Reindl Harald 16 Feb '15

16 Feb '15

Am 16.02.2015 um 22:53 schrieb Vadim Pestovnikov: > Thank you Reindl. > > So basically what that means is that mariadb is not 100% compatable with > percona and mysql? define 100% - replication is a *binary log* if it would be identical in every aspect it would exist only one product and even if all docs pretend 100% compatibility i would never come to the idea even run different minor versions in a replication setup just follow these instructions leads to the conclusion mix different products is a bad idea: http://dev.mysql.com/doc/refman/5.0/en/replication-upgrade.html P.S.: i don't appreciate off-list replies > On Feb 16, 2015 1:23 PM, "Reindl Harald" <h.reindl(a)thelounge.net > <mailto:h.reindl@thelounge.net>> wrote: > > > > Am 16.02.2015 um 22:19 schrieb Vadim Pestovnikov: > > Error in Log_event::read_log_event(): 'Found invalid event in binary > log', data_len: 44, event_type: 33 > > Error reading relay log event: slave SQL thread aborted because > of I/O error > > My slave is MariaDB 10 with data and logs partitions encrypted > by LUKS > and with SSL enabled. As far as RDS and Percona servers don't > use SSL I > am using non SSL user for the replication. > > Any recommendations or suggestions would be extremely appreciated > > > don't mix different server versions in replication setups > > there is a good reason why it is recommended always to update first > the slaves, run "mysql_upgrade" and then the masters to be 100% sure > that the maste rnever don't write anything into the binlog the slave > won't understand because differences in details

2 1

[Maria-discuss] MariaDB 10 multisource replication from AWS MySQL 5.6 RDS and Percona Server 5.6
by Vadim Pestovnikov 16 Feb '15

16 Feb '15

Hi all, I am trying to setup the multisource replication from AWS MySQL 5.6 RDS to MariaDB 10 using multisource replication. The installation was successful I can see from show slave 'connection' status\G that binary logs and their positions are changing, but If I add a new data on a master it is not replicated to the slave. Also tried to replicate from Percona Server 5.6 to MariaDB 10 using the multisource replication, but got errors like: Error in Log_event::read_log_event(): 'Found invalid event in binary log', data_len: 44, event_type: 33 Error reading relay log event: slave SQL thread aborted because of I/O error My slave is MariaDB 10 with data and logs partitions encrypted by LUKS and with SSL enabled. As far as RDS and Percona servers don't use SSL I am using non SSL user for the replication. Any recommendations or suggestions would be extremely appreciated. -- Vadim

2 1

[Maria-discuss] New Question: reading dbase records marked for deletion
by AskMonty KB 16 Feb '15

16 Feb '15

Hello, A new question has been asked in "Data Manipulation" by williamblocher. Please answer it at http://mariadb.com/kb/en/reading-dbase-records-marked-for-deletion/ as the person asking the question may not be subscribed to the mailing list. -------------------------------- Can you tell me if dbase records marked for deletion are treated as orinary records or ignored? Thank You Bill Blocher -------------------------------- To view or answer this question please visit: http://mariadb.com/kb/en/reading-dbase-records-marked-for-deletion/

1 0