June 2024 - discuss - lists.mariadb.org

External MyISAM partitions
by Jan Křístek 07 Aug '24

07 Aug '24

Hi, We have an old instance of MariaDB 10.3.9/10.3.39, with historical data. Every month I copy an old MyISAM partition (binary files) from the same version of the engine (but a different instance) to a big volume and replace an empty partition in the database with symlinks to the newly copied two files (MYD, MYI - when the engine is stopped). I have been doing it for a year or two, checking each time the data are accessible afterwards. This month it happened, that the data are not accessible anymore, and the database throwing an exception: *Got error 140: "Wrong create options" from storage engine MyISAM.* I tried doing it in a new instance, each time I replace a partition with a link to file elsewhere, the error appeared again. Does anyone have any ideas, why it's showing the error and how to avoid it? I cannot recall changing anything in the database configuration, upgrading any database, or doing any other changes. Thanks for any help you can provide, Jan

4 7

is blocked because of many connection
by Marc 02 Aug '24

02 Aug '24

I am constantly getting abuse from azure, amazon, digital ocean cloud. With previous mysql versions I never got this message. I assume you have set this to a default. I don't want to have this limit on ip but on ip+database, how can I change this?

6 39

MyISAM packed tables is uncompressed when index is added.
by Vassilis Virvilis 02 Jul '24

02 Jul '24

Hi, The question: Is uncompressing a myisam table when adding an index the expected behavior? Here is the story: I have a very big table so I thought to compress it. I used myisampack which worked great. The table.MYD shrank in size. myisamchk -ddv shows the table as compressed. I noticed the documentation that urges to run: myisamchk -rq after myisampack and I thought: Maybe I shouldn't create any indices on the uncompressed table since myisamchk will redo the work. But if I try to add an index to the compressed table suddenly it gets expanded to its original size and myisamchk -ddv does report the table a uncompressed. So the only solution is to: 1. add the indices in the uncompressed table 2. compress the table 3. run myisamchk -rq [--sort-index --analyze --parallel-recover --tmpdir=/somewhere/with/a/lot/of/space] 4. mysqladmin flush-tables; But this duplicates the index creations step. So is this the expected behavior? * If yes: why? Expanding the table behind the user's back is surprising... * If no: Should I report it as a bug? Here is how to replicate: #DROP TABLE IF EXISTS txxx; CREATE TABLE txxx (id INT, value TEXT); INSERT INTO txxx SELECT 1, REPEAT('A', 65536); *# Let's go to shell *ls -ltr txxx.* -rw-rw---- 1 mysql mysql 466 Jun 30 19:34 txxx.frm -rw-rw---- 1 mysql mysql 1024 Jun 30 19:40 txxx.MYI -rw-rw---- 1 mysql mysql *65548* Jun 30 19:40 txxx.MYD <-- 64K cool *# cool! let's compress *myisampack -f txxx Compressing txxx.MYD: (1 records) - Calculating statistics - Compressing file 87.42% ls -ltr txxx.* -rw-rw---- 1 mysql mysql 466 Jun 30 19:34 txxx.frm -rw-rw---- 1 mysql mysql *8255* Jun 30 19:40 txxx.MYD <-- 8K cool -rw-rw---- 1 mysql mysql 1024 Jun 30 19:41 txxx.MYI # great it is compressed *# because documentation says so: *myisamchk -qr txxx - check record delete-chain - recovering (with keycache) MyISAM-table 'txxx' Data records: 1 ls -ltr txxx.* -rw-rw---- 1 mysql mysql 466 Jun 30 19:34 txxx.frm -rw-rw---- 1 mysql mysql *8255* Jun 30 19:40 txxx.MYD <-- Still compressed -rw-rw---- 1 mysql mysql 1024 Jun 30 19:43 txxx.MYI myisamchk -ddv txxx MyISAM file: txxx Record format: *Compressed* <-------------------------------------!!!! OK Character set: utf8mb3_general_ci (33) File-version: 1 Creation time: 2024-06-30 19:40:36 Recover time: 2024-06-30 19:43:52 Status: checked Checksum: 2613455662 Data records: 1 Deleted blocks: 0 Datafile parts: 1 Deleted data: 0 Datafile pointer (bytes): 6 Keyfile pointer (bytes): 3 Datafile length: 8248 Keyfile length: 1024 Max datafile length: 281474976710654 Max keyfile length: 17179868159 Recordlength: 15 table description: Key Start Len Index Type Rec/key Root Blocksize ro *# Now Let's add the index* ALTER TABLE txxx ADD INDEX(id); *# Shell again *ls -ltr txxx.* -rw-rw---- 1 mysql mysql 958 Jun 30 19:47 txxx.frm -rw-rw---- 1 mysql mysql *65548* Jun 30 19:47 txxx.MYD <--- *Oups - now it is uncompressed* -rw-rw---- 1 mysql mysql 2048 Jun 30 19:47 txxx.MYI myisamchk -ddv txxx MyISAM file: txxx Record format: *Packed* <----------------------------- *Why*? Character set: utf8mb3_general_ci (33) File-version: 1 Creation time: 2024-06-30 19:47:30 Status: changed Data records: 1 Deleted blocks: 0 Datafile parts: 1 Deleted data: 0 Datafile pointer (bytes): 6 Keyfile pointer (bytes): 6 Datafile length: 65548 Keyfile length: 2048 Max datafile length: 281474976710654 Max keyfile length: 288230376151710719 Recordlength: 16 table description: Key Start Len Index Type Rec/key Root Blocksize 1 2 4 multip. long NULL 0 1024 1024

2 2

Can Mariabackup import from a Percona Xtrabackup backup?
by Bryan K. Walton 18 Jun '24

18 Jun '24

Hi, We have a Maria DB 5.5 replica set running on Centos 7. With this OS version about to become end of life, we are planning an upgrade of our replica set to RHEL 9 and Maria DB 10.5. My general plan is to add a new replica secondary to the mix, running Maria DB 10.5 and then promote that server to be primary and replace the other replica members with a new secondary that is also running Maria DB 10.5. My question is, since Mariabackup didn't exist in Maria DB 5.5, is it possible to backup our existing Maria databases with Percona Xtrabackup and then have Mariabackup (from Maria DB 10.5) import that backup on the new server? Thanks, Bryan

2 2

No repo access since two days
by mail＠alexander-palm.de 09 Jun '24

09 Jun '24

Hello there! Sorry, I have no idea where to post this, I hope it's the right place. Since two days I get 403 forbidden errors by apt update at all URLs from mariadb.com I try to get out the reason but I can't. What I have done: - check the trusted.d keys and reinstall it - check if the VHost provider is the problem. So I install a clean new VM at Strato with Ubuntu 22.04 and 24.04 and get the same error on both by using the 'mariadb_repo_setup' script. - install a clean VM at Hetzner whit ubuntu 24.04 and get the same problem by using the 'mariadb_repo_setup' script. I have no Idee what's happend. If I use the URL via curl there is no error. But I can see that the Request to https://dlm.mariadb.com/repo/mariadb-server/11.4/repo/ubuntu/dists/jammy/In… was redirectet to some cloudflare services with some tokens and timestamps and so on perhaps apt cannot handle that? 3 days before there was no issue with this configuration. Here is a output-example: bash mariadb_repo_setup # [info] Checking for script prerequisites. # [info] MariaDB Server version 11.4 is valid # [info] Repository file successfully written to /etc/apt/sources.list.d/mariadb.list # [info] Adding trusted package signing keys... # [info] Running apt-get update... E: The repository 'http://downloads.mariadb.com/Tools/ubuntu jammy InRelease' is not signed. E: Failed to fetch http://downloads.mariadb.com/Tools/ubuntu/dists/jammy/InRelease 403 Forbidden [IP: 2606:4700::6812:8718 80] E: Failed to fetch https://dlm.mariadb.com/repo/mariadb-server/11.4/repo/ubuntu/dists/jammy/In… 403 Forbidden [IP: 2606:4700::6811:bf0e 443] E: The repository 'https://dlm.mariadb.com/repo/mariadb-server/11.4/repo/ubuntu jammy InRelease' is not signed. E: Failed to fetch https://dlm.mariadb.com/repo/maxscale/latest/apt/dists/jammy/InRelease 403 Forbidden [IP: 2606:4700::6811:bf0e 443] E: The repository 'https://dlm.mariadb.com/repo/maxscale/latest/apt jammy InRelease' is not signed. # [error] Failed to add trusted package signing keys Anyone an idea? Thanks, Alex

1 1

Deleted database still present on replica
by Dave Hall 06 Jun '24

06 Jun '24

Hello, If this is not the right place for this question, please feel free to redirect... I am running MariaDB version 10.5.23 in a master/replica configuration with 4 replicas. I recently experienced a failure that required me to restore the DBs on all of my DB systems from backup, including all of my replicas. Somehow in the course of this process I have some databases on my some of my replicas that have been deleted on the master. What is the best way to remove these without confusing or breaking the replication process? Thanks. -Dave -- Dave Hall Binghamton University kdhall(a)binghamton.edu

2 1