- discuss - lists.mariadb.org

RAM/Swap Usage
by roughnecks 14 Sep '23

14 Sep '23

Hello, first time here :) I've got a single db for roundcube and yet after every couple days my swap for mariadb reaches 200MB and over. To be noted that I'm the sole roundcube's user and that even myself prefer using thunderbird, so I only access roundcube from time to time. Is this normal? I'm running debian stable, upgraded from oldstable when it came out and I remember that in bullseye ram usage was just 50MB or so. Thanks, have a nice day. -- roughnecks

3 10

Increasing RAM usage until oom
by Marco Dickert - evolver group 07 Sep '23

07 Sep '23

Hi folks, we experience a RAM issue with MariaDB (version 10.5.19-MariaDB-0+deb11u2-log) on a standard Debian bullseye system. The problem is that over time MariaDB uses more and more RAM, until the kernel's oom-killer terminates it. The server is a VMware guest with 8G RAM, 2G swap and 4 cores. The most important parameter, innodb_buffer_pool_size, is set to 2G. I know this is quite low for this system, but I want to have proof that MariaDB consumes defininitely more RAM than allowed. According to the calculation in this comment [1], the RAM usage shouldn't exceed 4.2G. Here is the result, including the raw configuration values: ``` MariaDB [(none)]> SELECT ROUND( -> ( @@GLOBAL.key_buffer_size -> + @@GLOBAL.query_cache_size -> + @@GLOBAL.tmp_table_size -> + @@GLOBAL.innodb_buffer_pool_size -> + @@GLOBAL.innodb_log_buffer_size -> + @@GLOBAL.max_connections * ( -> @@GLOBAL.sort_buffer_size -> + @@GLOBAL.read_buffer_size -> + @@GLOBAL.read_rnd_buffer_size -> + @@GLOBAL.join_buffer_size -> + @@GLOBAL.thread_stack -> + @@GLOBAL.binlog_cache_size) -> ) / 1024 / 1024, 1) `total MB`; +----------+ | total MB | +----------+ | 4220.0 | +----------+ 1 row in set (0,001 sec) MariaDB [(none)]> SELECT @@GLOBAL.key_buffer_size, @@GLOBAL.query_cache_size , @@GLOBAL.tmp_table_size, @@GLOBAL.innodb_buffer_pool_size , @@GLOBAL.innodb_log_buffer_size, @@GLOBAL.max_connections , @@GLOBAL.sort_buffer_size, @@GLOBAL.read_buffer_size , @@GLOBAL.read_rnd_buffer_size, @@GLOBAL.join_buffer_size , @@GLOBAL.thread_stack, @@GLOBAL.binlog_cache_size \G 1 row in set (0,000 sec) *************************** 1. row *************************** @@GLOBAL.key_buffer_size: 10485760 @@GLOBAL.query_cache_size: 536870912 @@GLOBAL.tmp_table_size: 16777216 @@GLOBAL.innodb_buffer_pool_size: 2147483648 @@GLOBAL.innodb_log_buffer_size: 134217728 @@GLOBAL.max_connections: 512 @@GLOBAL.sort_buffer_size: 2097152 @@GLOBAL.read_buffer_size: 131072 @@GLOBAL.read_rnd_buffer_size: 262144 @@GLOBAL.join_buffer_size: 262144 @@GLOBAL.thread_stack: 299008 @@GLOBAL.binlog_cache_size: 32768 ``` However, MariaDB uses up to 5.4G so far. And it keeps increasing. Last time the oom-killer was invoked, MariaDB used 10G total, and 7.4G rss (given in 4k pages): ``` [Mi Jul 26 16:01:06 2023] Tasks state (memory values in pages): [Mi Jul 26 16:01:06 2023] [ pid ] uid tgid total_vm rss pgtables_bytes swapents oom_score_adj name ... [Mi Jul 26 16:01:06 2023] [3462912] 10000 3462912 2753196 1895256 20762624 472509 0 mariadbd ... [Mi Jul 26 16:01:06 2023] oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0,global_oom,task_memcg=/system.slice/mariadb.service,task=mariadbd,pid=3462912,uid=10000 [Mi Jul 26 16:01:06 2023] Out of memory: Killed process 3462912(mariadbd) total-vm:11012784kB, anon-rss:7581024kB, file-rss:0kB,shmem-rss:0kB, UID:10000 pgtables:20276kB oom_score_adj:0 ``` The MariaDB knowledge base [2] says that the problem probably correlates with one of the following options: ``` +----------------------------+----------------------+----------------+ | variable_name | default_value | variable_value | +----------------------------+----------------------+----------------+ | INNODB_IO_CAPACITY_MAX | 18446744073709551615 | 2000 | | QUERY_CACHE_LIMIT | 1048576 | 67108864 | | MAX_CONNECTIONS | 151 | 512 | | TABLE_OPEN_CACHE_INSTANCES | 8 | 7 | | LONG_QUERY_TIME | 10.000000 | 5.000000 | | KEY_BUFFER_SIZE | 134217728 | 10485760 | | QUERY_CACHE_SIZE | 1048576 | 536870912 | | INNODB_LOCK_WAIT_TIMEOUT | 50 | 80 | | INNODB_LOG_BUFFER_SIZE | 16777216 | 134217728 | | HOST_CACHE_SIZE | 128 | 654 | | INNODB_FT_MIN_TOKEN_SIZE | 3 | 2 | | WAIT_TIMEOUT | 28800 | 600 | | INNODB_FAST_SHUTDOWN | 1 | 0 | | INNODB_LOG_FILE_SIZE | 100663296 | 1073741824 | | THREAD_POOL_SIZE | 8 | 4 | | INNODB_BUFFER_POOL_SIZE | 134217728 | 2147483648 | | MAX_ALLOWED_PACKET | 16777216 | 268435456 | | BACK_LOG | 150 | 254 | +----------------------------+----------------------+----------------+ ``` I can't see what setting could probably cause the issue. Some values are Debian defaults. Most of them we set intentionally. Can anyone give a hint on where to look for? Do you need further information? Thanks for any help. [1] https://serverfault.com/a/1020847 [2] https://mariadb.com/kb/en/mariadb-memory-allocation/#how-to-troubleshoot-ou… Regards, Marco

3 4

Odp: Re: Re: Rocksdb as a replacement for toku?
by pslawek83 25 Aug '23

25 Aug '23

Hi Marko, thanks for the response ... so for the DOCS part > h mariadb.com ttps://mariadb.com/kb/en/innodb-page-compression/ > When InnoDB page compression is used, InnoDB may still write the compressed page to the tablespace file with the original size of the uncompressed page, which would be equivalent to the value of the mariadb.com innodb_page_size system variable. This is done by design, because when InnoDB's I/O code needs to read the page from disk, it can only read the full page size. However, this is obviously not optimal. > This punch hole technique allows InnoDB to read the compressed page from disk as the full page size, even though the compressed page really takes up less space on the file system. So as far as i understand if compression is enabled the server will write eg. 40% utilized page which will later be decompressed when read, but this compression/decompression won't have any advangage if there's no punch hole support. The only difference will be that the block will be filled with zeros for the empty space at the end. Or is it possible for PAGE_COMPRESSION to still be used without punch hole (just not that effective for example)? Also tested this row compression, and yes page compression seems much better. Is it possible to get rid of punch hole by just re-mapping the pages inside these innodb files? I was thinking about this... if it's possible to change the data structure to divide the file into 512b blocks while the beginning of each block will mark page number (0xffffffff for blocks which are part of previous page). To increase page size we'd need to add another page with same number at the end of the file, later continuous recycled blocks can be added to a list and used to store smaller pages. To make this atomic, first whole block would be allocated and written with 0xfffffffe as a dummy id and then this placeholder id could be replaced with real block number). The problem with this could be that the file would get 100% fragmented over time (but probably will work ok with NVME). The mapping could be read back by reading whole datafile during server start and it'd need 125MB of memory for 1TB of compressed data to store the mapping in memory for fast access. What you thinik? Would it be easy to add it on top of current code (if there's any API maybe, so for example i can do this) and would it even make sense? Best, Slawomir. Dnia 21 sierpnia 2023 07:08 Marko Mäkelä < marko.makela(a)mariadb.com > napisał(a): On Sun, Aug 20, 2023 at 6:12 PM pslawek83 via discuss <discuss(a)lists.mariadb.org> wrote: Thanks everyone for the answers. Tested this PAGE_COMPRESSION a bit and make some reading. Not sure if I fully understand how that works... is that correct? 1. Impossible without punch hole as if we don't have it the server will just write partially used pages which will take full page size anyway. But in docs it seems it described its use with punch hole disabled. What's the reason to still have page compression without punch hole? Can you please point to the misleading part of the documentation so that it can be corrected? 2. After enabling compression each page needs to be trimmed using punch hole, so if we have 1 TB database and 64k blocks, the database files will have over 15 million fragments (1000000000000/64000) That is correct. So if we get like 50% compression and then we'll be able to put another 1TB of data into these holes we'll be having close to 31 million file fragments. I'm not sure if under such conditions the FS will be able to even allocate space efficiently anymore, copy these files, etc. Or im not getting something? For what it is worth, around the same time when the page_compressed was being developed for MariaDB Server 10.1, something similar was being worked on at Oracle, I think for MySQL 5.7. To my understanding, the same hardware partner worked with both companies. The obvious answer to smalldatum.blogspot.com https://smalldatum.blogspot.com/2015/10/wanted-file-system-on-which-innodb.… would have been en.wikipedia.org https://en.wikipedia.org/wiki/Fusion-io_NVMFS if it had worked without kernel panics or existed in a mainline kernel. I don’t even know if it ever was open sourced. 3. I was testing compression of ~1.5GB table with nice results on default 16kb page size. One thing i noticed is that l was always getting exactly the same compressed size using `du`. However the table read/copy times were different from 40s for uncompressed to 80-400 seconds depending on algo. Is that possible that is because page size is too low? As far as I remember, most compression algorithms that are derivatives of the LZ77 family have an input buffer size of 32 KiB. From that point of view, it would seem to make sense to use innodb_page_size=32k or innodb_page_size=64k instead of the default. Furthermore, I think that it was a mistake to implement bzip2 support at all. That algorithm would work with much larger buffer sizes. 4. Myrocks seems to be very slow compared to compressed inno. Was also reading docs related to it. For its usecase is it more suitable for storing large rows rather than many small rows? To my understanding, the LSM trees that MyRocks is built on work best in an insert-only workload. On top of LSM trees, MyRocks implements some caches and Bloom filters. If you have few updates or deletes and most reads are accessing recently inserted data, it could work. 5. I remember reading that some innodb compression is going to be obsolete. Is that row compression? Or there are no plans to obsolete any compression model? I wanted to deprecate and remove the ROW_FORMAT=COMPRESSED that I designed and implemented in the InnoDB Plugin for MySQL 5.1 based on some high level ideas of Heikki Tuuri between 2005 and 2007. jira.mariadb.org https://jira.mariadb.org/browse/MDEV-22367 was the first step towards that. Based on community feedback, this won’t be happening. Even after some data structure cleanup that took place in MariaDB Server 10.6, supporting the format incurs an overhead of some 32 to 40 bytes per buffer page descriptor, which is not insignificant. Removing both ROW_FORMAT=COMPRESSED and the adaptive hash index would shrink the block descriptor to 80 bytes from its current 10.6 size of 160 bytes. The write performance with ROW_FORMAT=COMPRESSED should have been degraded when I implemented the easier-to-parse log record format in MDEV-12353 (MariaDB Server 10.5). There are open bugs about ROW_FORMAT=COMPRESSED, such as jira.mariadb.org https://jira.mariadb.org/browse/MDEV-31574. I would say that the InnoDB ROW_FORMAT=COMPRESSED seemed like a good idea back when it was implemented, when HDD storage was the only practical option. Just like the InnoDB change buffer, which I removed in MariaDB Server 11.0 after having to deal with too many cases of corruption: fosdem.org https://fosdem.org/2023/schedule/event/innodb_change_buffer/ With best regards, Marko -- Marko Mäkelä, Lead Developer InnoDB MariaDB plc

3 3

Re: Rocksdb as a replacement for toku?
by pslawek83 22 Aug '23

22 Aug '23

Thanks everyone for the answers. Tested this PAGE_COMPRESSION a bit and make some reading. Not sure if I fully understand how that works... is that correct? 1. Impossible without punch hole as if we don't have it the server will just write partially used pages which will take full page size anyway. But in docs it seems it described its use with punch hole disabled. What's the reason to still have page compression without punch hole? 2. After enabling compression each page needs to be trimmed using punch hole, so if we have 1 TB database and 64k blocks, the database files will have over 15 million fragments ( 1000000000000/64000) So if we get like 50% compression and then we'll be able to put another 1TB of data into these holes we'll be having close to 31 million file fragments. I'm not sure if under such conditions the FS will be able to even allocate space efficiently anymore, copy these files, etc. Or im not getting something? 3. I was testing compression of ~1.5GB table with nice results on default 16kb page size. One thing i noticed is that l was always getting exactly the same compressed size using `du`. However the table read/copy times were different from 40s for uncompressed to 80-400 seconds depending on algo. Is that possible that is because page size is too low? 4. Myrocks seems to be very slow compared to compressed inno. Was also reading docs related to it. For its usecase is it more suitable for storing large rows rather than many small rows? 5. I remember reading that some innodb compression is going to be obsolete. Is that row compression? Or there are no plans to obsolete any compression model? Thanks. Dnia 16 sierpnia 2023 16:08 Gordan Bobic via discuss < discuss(a)lists.mariadb.org > napisał(a): Larger pages will yield higher compression ratios, as will zstd (slightly slower than lz4). Mileage varies, but on some of my databases, between the 64KB pages (zfs recordsize) and zstd, my compression ratios approximately doubled, from about 1.8:1 to a little under 4:1. This will all heavily depend on the type of data you are storing, though. zstd has adjustable compression, so if you have tons of CPU to spare, you could turn it up from the default of 3 (probably optimal cost/size compromise) to high teens for a compression boost. On Wed, 16 Aug 2023, 14:34 Reinis Rozitis, < r(a)roze.lv > wrote: > From: Gordan Bobic via discuss < discuss(a)lists.mariadb.org > > > One of the best compromises is to configure InnoDB with 64KB pages, put it on ZFS with recordsize=64K, and enable zstd compression on ZFS. Sorry for hijacking the thread a bit. After the discontinuation of tokudb I chose the innodb+zfs route (bad previous/early experience with rocks, also million knobs to adjust for your workload etc) but at that time didn't think of increasing the Innodb page size to 64KB - do you have any (even rough) numbers of 16 vs 64KB (on zfs)? Also, what would be the benefit of zstd vs lz4? Higher compression ratio? rr _______________________________________________ discuss mailing list -- discuss(a)lists.mariadb.org To unsubscribe send an email to discuss-leave(a)lists.mariadb.org

3 2

Running events on same time
by kpesio＠gmail.com 17 Aug '23

17 Aug '23

Hi. Could someone confirm in some way about running multiple events at same time. Documentation at https://mariadb.com/kb/en/automating-mariadb-tasks-with-events/ states: "The event scheduler runs as a single thread. If any event is supposed to start while another event is running, its execution will be skipped." I have simulated events triggering at same time, running with delays to get all kinds of overlapping schedules, writing to same table from the event and not missed a single one. The information_schema.events tables column last_executed has - depending on scheduled time - valid values and I can see in the processlist events running at same time. Version I'm running is 10.6.10 on Linux. Tried to cross check against the MySQL documentation but couldn't find anything about this there. So what situation does the documentation refer to or it this something fixed long time ago?

3 5

Rocksdb as a replacement for toku?
by pslawek83 16 Aug '23

16 Aug '23

Hi Guys, i'm looking for tokudb replacement and i'm again thinking about rocksdb. Is it mature enough to be used as an engine for statistical tables (several billions rows, around 2TB in size?). - Is it (will it be?) possible to have something like single-file-per-table (or single dir per table?) - Is it shared per all tables, so all tables are still using single set of files (can single crashed table/row lead to all tables being unusable?) - Will partitioning the table have performance benefits like with other engines? Can I do instant TRUNCATE PARTITION for easy data retention? Thanks

5 6

Query on Building Maria DB for different Compression Algorithm
by Raj, Rahul 10 Aug '23

10 Aug '23

[AMD Official Use Only - General] Hi, I am building MariaDB from source code and have queries on linking of compression algorithm. Before building the source code of MariaDB, I am exporting the path for SNAPPY or LZMA or the algorithm which is currently it supports. Then I build the code with - cmake -B . -DCMAKE_INSTALL_PREFIX=$PWD/maria_bin -DPLUGIN_MROONGA=NO -DPLUGIN_ROCKSDB=NO cmake --build . -j cmake -install . It compiled the code and installed in the specified location. When I checked the lib/plugin folder it has generated, provider_lzma.so or provider_snappy.so files and when I did ldd on that .so file, it has given the path from where it is fetching the liblzma.so.1 or libsnappy.so.1 file. Query: 1. The path which ldd shows for provider_lzma.so or provider_snappy.so is taking form default path, and it is not from the path which I am exporting the library and bin path - could you please explain me the correct steps. 2. Is there any way or any Flag we can set to switch between different compression method during build time itself, and in which log we can check it has linked properly. 3. How to benchmark the linked compression method and how we can see it is using the linked compression method from the custom path of .so files. Kindly help on above query. Thanks Rahul Raj

4 9

Reduce logging output from mariabackup
by Tom Worster 03 Aug '23

03 Aug '23

Is there an option to reduce the volume of information mariabackup writes to the console as it does its magic? Some of the log lines are very welcome, for example I like the first three Aug 03 11:35:03 bhs6 v2_backup_maria.sh[1291066]: [00] 2023-08-03 11:35:03 Connecting to MariaDB server host: localhost, user: mariabackup, password: set, port: not set, socket: /run/mysqld/mysqld.sock Aug 03 11:35:03 bhs6 v2_backup_maria.sh[1291066]: [00] 2023-08-03 11:35:03 Using server version 10.6.14-MariaDB-1:10.6.14+maria~deb11-log Aug 03 11:35:03 bhs6 v2_backup_maria.sh[1291066]: mariabackup based on MariaDB server 10.6.14-MariaDB debian-linux-gnu (x86_64) and the last two are really great, like a gift from god Aug 03 11:36:34 bhs6 v2_backup_maria.sh[1291196]: [00] 2023-08-03 11:36:34 Last binlog file ./bhs8-bin.001617, position 952620696 Aug 03 11:36:34 bhs6 v2_backup_maria.sh[1291196]: [00] 2023-08-03 11:36:34 completed OK! But there's about 500 between these every hour in the journal and most of them are quite boring like this Aug 03 11:36:31 bhs6 v2_backup_maria.sh[1291066]: [01] 2023-08-03 11:36:31 Copying ./sys/sys_config.frm to /pool2/mariabackup/sys/sys_config.frm Aug 03 11:36:31 bhs6 v2_backup_maria.sh[1291066]: [01] 2023-08-03 11:36:31 ...done Or stuff I do not understand and suspect I probably don't need to undersand like Aug 03 11:35:03 bhs6 v2_backup_maria.sh[1291066]: [00] 2023-08-03 11:35:03 uses posix_fadvise().

3 5

Expected release date MariaDB 10.6.15
by William Edwards 01 Aug '23

01 Aug '23

Hi, Is MariaDB 10.6.15 expected to be released this week? According to Jira[1], it is 4 days overdue. This version will include a fix that I'm waiting for. Met vriendelijke groeten, William Edwards [1]: https://jira.mariadb.org/projects/MDEV/versions/29013

3 4

Help us redesign the mariadb.org website
by Ian Gilfillan 17 Jul '23

17 Jul '23

We are currently re-designing the mariadb.org website and you are invited to participate in this process. The interactive study shouldn't take longer than 10 to 15 minutes to complete and will close at midnight on Thursday 20 July 2023 (UTC-8). Access the study here: (no registration or login is required) https://study.uxtweak.com/cardsort/gkeYA0HPSXawp6YNsjllU The activity involves a method called Card Sorting, which will influence how we design the information architecture of the website: its structure, navigation and the findability of content and functionality. Your inclusion in this effort is important and will help to inform and assist us in designing an intuitive, usable and effective future website. Many thanks from the website redesign team MariaDB Foundation

1 0

Advice on Replication
by Andrew Gable 13 Jul '23

13 Jul '23

5 4

Re: Sandboxing mariabackup
by Tom Worster 07 Jul '23

07 Jul '23

Hi Daniel, That part is looking good for now but I gotta do fire drills validate the whole shebang. There's something almost blog worthy about this. mysqldump started as a handy import export utility and ended up being an exquisite point-in-time logical SQL snapshot tool. Now xtrabackup and mariabackup are so astonishingly fast at restoration/recovery they can't be ignored. But they are terrifying! They operate directly on the datadir. mysqldump is a nice friendly SQL client like any other. Does that make sense? An SQL client can ruin the DB as well as anything else if it has the right permissions (pun!). Hence subtracting permission to write datadir from the mariabackup process feels like a step to parity. Tom On 7/6/2023 6:52:17 PM, "Daniel Black" <daniel(a)mariadb.org> wrote: >Glad you worked it out. > >I did mean to write this onlist a while ago. Got saved a draft message >and I forgot about it. > >On Thu, Jul 6, 2023 at 10:47 PM Tom Worster <fsb(a)thefsb.org> wrote: >> >> On 7/6/2023 7:14:50 AM, "Daniel Black" <daniel(a)mariadb.org> wrote: >> >> >> I prefer to sandbox systemd services with ProtectSystem=strict. >> >> Especially so when User=root. >> > >> >Should be able to use User=mysql (or whatever the mariadb.service is >> >running as), provided the backup directory is also mysql: owned. >> > >> Thanks Daniel. >> >> Yes. I discovered that in subsequent testing. >> spin >> And I found that only target-dir and tmpdir need to be read-write. I >> forget what made me think that wasn't enough. >> >> To run as mariabackup as mysql user but with read-only access to datadir >> I separated the mariabackup part of the overall hourly backup script >> into its own service unit: >> [Unit] >> Description=mariabackup DBs >> OnFailure=unit-status-mail(a)%n.service unit-status-pushover(a)%n.service >> [Service] >> User=mysql >> Type=oneshot >> ProtectSystem=strict >> ReadWritePaths=/pool2/mariabackup /pool2/mariabackup_tmpdir >> ExecStart=/path/to/backup_maria.sh >> >> This is /path/to/backup_maria.sh. It's on a replica and the test of prev >> and new size are a check (among others) that the replica isn't stuck. >> #!/bin/bash >> prev_backup_size=`du -s /pool2/mariabackup` >> rm -rf /pool2/mariabackup/* >> mariabackup --backup --target-dir=/pool2/mariabackup/ >> --tmpdir=/pool2/mariabackup_tmpdir --user=mariabackup --password=__ >> mariabackup --prepare --target-dir=/pool2/mariabackup/ >> --tmpdir=/pool2/mariabackup_tmpdir >> new_backup_size=`du -s /pool2/mariabackup` >> if [ "$prev_backup_size" = "$new_backup_size" ] >> then >> echo WARNING: Consecutive mariabackup sizes unchanged >> # FIXME send alerts >> fi >>

2 2

Sandboxing mariabackup
by Tom Worster 06 Jul '23

06 Jul '23

I prefer to sandbox systemd services with ProtectSystem=strict. Especially so when User=root. So when ProtectSystem=strict, which dirs need to be ReadWritePaths to run `mariabackup --backup` and `mariabackup --prepare`? So far I got the --target-dir and the --tmpdir. Is that sufficient? tia Tom

2 2

stop database before truncate
by robert k Wild 03 Jul '23

03 Jul '23

hi all, looking at this guide https://www.ibm.com/support/pages/temporarily-reduce-space-taken-console-da… do i ideally have to stop the database before i truncate some tables ie /opt/aspera/common/mysql/bin/mysql -u root -p'password' aspera_console TRUNCATE* fasp_sessions fasp_files fasp_rates * *jobs job_steps* *thanks,* *rob* -- Regards, Robert K Wild.

2 5

Can we place a temporary table inside a stored function
by manpritsinghece＠gmail.com 29 Jun '23

29 Jun '23

Just need to know if we can place a temporary table inside the stored function ?

3 6

can you purge data older than 4 months old
by robert k Wild 29 Jun '23

29 Jun '23

hi all, i was just wondering i know about the "mysqlcheck -a and -o" ie to optimize and to analyse ie to cut it down in size but NOT to purge any data but is there like a purge option to delete all data older than 4 months old in the database thanks, rob -- Regards, Robert K Wild.

2 2

2023 MariaDB (Un)Conference and MariaDB Server Fest
by Ian Gilfillan 28 Jun '23

28 Jun '23

The MariaDB (Un)Conference, for MariaDB Server developers, and the MariaDB Server Fest, for anyone interested in MariaDB Server, will be taking place from 3 - 5 October as hybrid events, both in Helsinki and online. See https://mariadb.org/fest2023/ for details. The CfP for both events is available at https://mariadb.org/fest2023cfp/

1 0

mysql restore error
by robert k Wild 26 Jun '23

26 Jun '23

hi all, i successfully backed up the database /opt/aspera/common/mysql/bin/mysqldump -u root -p'password' aspera_console > /mnt/pixit/engineering/aspera_console/mysql_backups/20230623.sql then i try to restore it, first i delete the database and create the database /opt/aspera/common/mysql/bin/mysql -u root -p'password' drop database aspera_console; create database aspera_console; then run the restore command and i get the error /opt/aspera/common/mysql/bin/mysql -u root -p'password' aspera_console < /mnt/pixit/engineering/aspera_console/mysql_backups/20230623.sql ERROR 1305 (42000) at line 80504: FUNCTION aspera_console.report_user_id does not exist luckily this is not the live vm, its a clone of the live, so im messing around thanks, rob -- Regards, Robert K Wild.

2 4

Very large ibdata1 growth (335Gb) with 10.6.13, solved in 10.6.14 ?
by Simon Avery 20 Jun '23

20 Jun '23

Hi All, We run around 60 Mariadb 10.6 servers on EL in vms and varying workloads. Some time after they updated from 10.6.12 to 10.6.13, four of the busiest started showing higher disk usage than expected. Upon investigation, /var/lib/mysql/ibdata1 was exceptionally large, up to 335Gb, and was still growing at several Gb/day. Note: All servers have only run with "innodb_file_per_table" set and only have innodb engines for user DBs. I cobbled together some data collection just before I went on holiday and checked the graphs today. (Pic attached - if not stripped by the list software) [A picture containing screenshot, line, parallel Description automatically generated] Four machines growing, rest static. Two affected machines stopped growing a week ago - and on checking, Mariadb was upgraded from 10.6.13 to 10.6.14 on them and they were rebooted then. The two still growing are still running 10.6.13 My questions! 1. Am I right that this was a known bug in Mariadb and has been resolved with the .14 release? 2. The .14 release notes mention "InnoDB does not free UNDO after the fix of MDEV-30671 (MDEV-31234)" - is the failure to free UNDO logs a likely cause for ibdata1 to grow, given that it stores the undo log? (The two worst affected machines process many million queries a day) 3. To recover this space, I will need to do a full mysqldump, wipe and reimport of the data, correct? I guess I'm mostly asking for reassurance that this issue is now resolved and that once I schedule in the downtime to recover the space, that it's not going to resurface... Thanks, Simon

2 2

Fwd: DBA in Tokyo
by James Tobin 19 Jun '23

19 Jun '23

Hello, I'm working with an employer that is looking to hire a DBA in Tokyo. Consequently, I had hoped some members of this list may like to discuss. I can be reached using "JamesBTobin (at) Gmail (dot) Com". Kind regards, James

1 0

Connector/C fails to read MariaDB 10.6+ binary logs
by sruli67＠gmail.com 15 Jun '23

15 Jun '23

Hi, I'm using MariaDB Connector/C to 3.3.5 to read binary logs. It works perfectly for MySQL 8.0 and MariaDB 10.5 databases. For MariaDB 10.6 and above, mariadb_rpl_fetch() crashes (!) if the binlog contains XA transactions. Specifically, the function fails when it tries to read event type GTID with this contents: 'XA START X\'54584e31\',X\'\',1 GTID 0-1-21' Please advise Sruli Ganor

4 9

[Maria-discuss] Backup on the replication server getting affected
by ragul rangarajan 09 Jun '23

09 Jun '23

3 3

Re: [Maria-discuss] Restore from mysqldump fails with "Error in foreign key constraint"
by mark f edwards 05 Jun '23

05 Jun '23

How is providers in the `xu6gc_tj_notification_providers' table defined? Is it also varchar(100) character sat uft8mbr collate utf8mb4-general-ci ?? The parent and child columns has to match exactly. On June 3, 2023, at 7:55 AM, Alex <mysqlstudent(a)gmail.com> wrote: Hi, I'm trying to use mysqlbackup to backup a mariadb-10.5.19 database on fedora37 server using the following: # mysqldump --verbose -u root -ppass --result-file=/var/backup/mysql-backup/backup-lsalex-2023-06-03-13008.sql lsalex I usually use a more involved mysqlbackup command, but it also failed, so I figured it would be better to start out more simply as above. It fails with the following when restoring to a new, test database: # mysql -u root -p lstest < /var/backup/mysql-backup/backup-lsalex-2023-06-03-13008.sql Enter password: ERROR 1005 (HY000) at line 12884: Can't create table `lsalex`.`xu6gc_tj_notification_user_exclusions` (errno: 150 "Foreign key constraint is incorrectly formed") This is a backup/restore of a Joomla database. Am I using the wrong command to perform the backup or restore? This isn't a database that I created and I really have no idea how it works. I've also searched extensively for this error, and it appears to be a programming error, not a database error, but obviously I have no control over Joomla programming. Here's what I found from "SHOW ENGINE INNODB STATUS": LATEST FOREIGN KEY ERROR ------------------------ 2023-06-03 10:09:01 0x7f3a8d8516c0 Error in foreign key constraint of table `lstest`.`xu6gc_tj_notification_user_exclusions`: Create table `lstest`.`xu6gc_tj_notification_user_exclusions` with foreign key `xu6gc_tj_notification_user_exclusions_ibfk_1` constraint failed. Field type or character set for column 'provider' does not match referenced column 'provider'.------------ Here's the lines from the backup mentioned in the error message: 12884 CREATE TABLE `xu6gc_tj_notification_user_exclusions` ( 12885 `user_id` int(11) NOT NULL, 12886 `client` varchar(100) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NOT NULL, 12887 `key` varchar(100) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NOT NULL, 12888 `provider` varchar(100) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NOT NULL, 12889 KEY `client1` (`client`,`provider`(50),`key`), 12890 KEY `key` (`key`), 12891 KEY `provider` (`provider`), 12892 CONSTRAINT `xu6gc_tj_notification_user_exclusions_ibfk_1` FOREIGN KEY (`provider`) REFERENCES `xu6gc_tj_notification_providers` (`provider`) 12893 ) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_general_ci; Any ideas are greatly appreciated.

5 4

[Maria-discuss] Restore from mysqldump fails with "Error in foreign key constraint"
by Alex 03 Jun '23

03 Jun '23

Hi, I'm trying to use mysqlbackup to backup a mariadb-10.5.19 database on fedora37 server using the following: # mysqldump --verbose -u root -ppass --result-file=/var/backup/mysql-backup/backup-lsalex-2023-06-03-13008.sql lsalex I usually use a more involved mysqlbackup command, but it also failed, so I figured it would be better to start out more simply as above. It fails with the following when restoring to a new, test database: # mysql -u root -p lstest < /var/backup/mysql-backup/backup-lsalex-2023-06-03-13008.sql Enter password: ERROR 1005 (HY000) at line 12884: Can't create table `lsalex`.`xu6gc_tj_notification_user_exclusions` (errno: 150 "Foreign key constraint is incorrectly formed") This is a backup/restore of a Joomla database. Am I using the wrong command to perform the backup or restore? This isn't a database that I created and I really have no idea how it works. I've also searched extensively for this error, and it appears to be a programming error, not a database error, but obviously I have no control over Joomla programming. Here's what I found from "SHOW ENGINE INNODB STATUS": LATEST FOREIGN KEY ERROR ------------------------ 2023-06-03 10:09:01 0x7f3a8d8516c0 Error in foreign key constraint of table `lstest`.`xu6gc_tj_notification_user_exclusions`: Create table `lstest`.`xu6gc_tj_notification_user_exclusions` with foreign key `xu6gc_tj_notification_user_exclusions_ibfk_1` constraint failed. Field type or character set for column 'provider' does not match referenced column 'provider'.------------ Here's the lines from the backup mentioned in the error message: 12884 CREATE TABLE `xu6gc_tj_notification_user_exclusions` ( 12885 `user_id` int(11) NOT NULL, 12886 `client` varchar(100) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NOT NULL, 12887 `key` varchar(100) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NOT NULL, 12888 `provider` varchar(100) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NOT NULL, 12889 KEY `client1` (`client`,`provider`(50),`key`), 12890 KEY `key` (`key`), 12891 KEY `provider` (`provider`), 12892 CONSTRAINT `xu6gc_tj_notification_user_exclusions_ibfk_1` FOREIGN KEY (`provider`) REFERENCES `xu6gc_tj_notification_providers` (`provider`) 12893 ) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_general_ci; Any ideas are greatly appreciated.

1 0

Re: [Maria-discuss] Backup on the replication server getting affected
by andrei.elkin＠pp.inet.fi 02 Jun '23

02 Jun '23

Howdy Ragul, > Hi Andrei, > > Do we have any procedures to reproduce the issue MDEV-30780? Thanks for posting the gdb bt:s. They rule out 30780 yet not suggesting to me enough about the hang reason. This is something new to me and does deserve filing an MDEV ticket. Still I'd defer that until one has confirmed the same issue is seen on the latest 10.6. So you could run your load against the most recent slave version that'd be at least the safest (for our time). It might be (a slave worker) Thread 80 spinning inside #6 0x000055de407a0a3c in log_write_up_to (lsn=<optimized out>, lsn@entry=216757233923297, flush_to_disk=flush_to_disk@entry=false, rotate_key=rotate_key@entry=false, a goto repeat "loop". That hopefully you can confirm any next time the hang appears back. Could you please check whether #6 calls iteratively indeed `group_commit_lock::release()`? (With e.g (gdb) br thd_decrement_pending_ops thread 80 of course the number may change:-)). All the other slave worker threads may be waiting for the 80 but I can't confirm that until more data gets available. Namely I need to see the output of (gdb) thr app all get_about_worker_thread where the latter is defined as define get_about_worker_thread if $_any_caller_is ("handle_rpl_parallel_thread", 50) bt p handle_rpl_parallel_thread::rpt if (handle_rpl_parallel_thread::rpt->thd->rgi_slave) p handle_rpl_parallel_thread::rpt->thd->rgi_slave p handle_rpl_parallel_thread::rpt->thd->rgi_slave->current_gtid p handle_rpl_parallel_thread::rpt->thd->rgi_slave->gtid_sub_id p handle_rpl_parallel_thread::rpt->thd->rgi_slave->worker_error end end end > Unable to reproduce the issue locally but it occurs at random. to require some more patience from us. I belive we can resolve it while you're helping so generously! Cheers, Andrei > > Regards, > Ragul > > On Mon, May 29, 2023 at 7:06 PM ragul rangarajan <ragulrangarajan(a)gmail.com> wrote: > > Thanks Andrei, > > Hope my issue is more related to the issue MDEV-30780 optimistic parallel slave hangs after hit an error > Trying to reproduce with a minimal database.

1 0

[Maria-discuss] FYI:10.5.15-MariaDB crash und recover
by Radisson97＠gmx.de 01 Jun '23

01 Jun '23

Hello I had a crash today, there was no problem with the hardware. Maybe someone is interessted or find some useful information. It is a production machine so no core file. have fun. btw: Linux 5.10.0-17-amd64 #1 SMP Debian 5.10.136-1 == mariadb.log == 230601 8:22:09 [ERROR] mysqld got signal 11 ; This could be because you hit a bug. It is also possible that this binary or one of the libraries it was linked against is corrupt, improperly built, or misconfigured. This error can also be caused by malfunctioning hardware. To report this bug, see https://mariadb.com/kb/en/reporting-bugs We will try our best to scrape up some info that will hopefully help diagnose the problem, but since we have already crashed, something is definitely wrong and this may fail. Server version: 10.5.15-MariaDB-0+deb11u1-log key_buffer_size=134217728 read_buffer_size=131072 max_used_connections=11 max_threads=252 thread_count=11 It is possible that mysqld could use up to key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 685818 K bytes of memory Hope that's ok; if not, decrease some variables in the equation. Thread pointer: 0x7faf440061d8 Attempting backtrace. You can use the following information to find out where mysqld died. If you see no messages after this, something went terribly wrong... stack_bottom = 0x7fafa439bd38 thread_stack 0x49000 ??:0(my_print_stacktrace)[0x55ebc2f0954e] ??:0(handle_fatal_signal)[0x55ebc2a08f65] sigaction.c:0(__restore_rt)[0x7fb11f670140] ??:0(_ma_state_info_write)[0x55ebc2bfdc73] ??:0(_ma_state_info_write)[0x55ebc2bff737] ??:0(_ma_state_info_write)[0x55ebc2c0746c] ??:0(handler::ha_write_row(unsigned char const*))[0x55ebc2a15aa8] ??:0(MyCTX_nopad::finish(unsigned char*, unsigned int*))[0x55ebc2c420da] ??:0(handler::ha_write_row(unsigned char const*))[0x55ebc2a15aa8] ??:0(write_record(THD*, TABLE*, st_copy_info*, select_result*))[0x55ebc27d587d] ??:0(mysql_insert(THD*, TABLE_LIST*, List<Item>&, List<List<Item> >&, List<Item>&, List<Item>&, enum_duplicates, bool, select_result*))[0x55ebc 27dd8a4] ??:0(mysql_execute_command(THD*))[0x55ebc2810716] ??:0(mysql_parse(THD*, char*, unsigned int, Parser_state*, bool, bool))[0x55ebc28145db] ??:0(dispatch_command(enum_server_command, THD*, char*, unsigned int, bool, bool))[0x55ebc2816a5d] ??:0(do_command(THD*))[0x55ebc28182de] ??:0(do_handle_one_connection(CONNECT*, bool))[0x55ebc2909fb2] ??:0(handle_one_connection)[0x55ebc290a22d] ??:0(MyCTX_nopad::finish(unsigned char*, unsigned int*))[0x55ebc2c4611b] nptl/pthread_create.c:478(start_thread)[0x7fb11f664ea7] x86_64/clone.S:97(__GI___clone)[0x7fb11f27bdef] Trying to get some variables. Some pointers may be invalid and cause the dump to abort. Query (0x7faf44014860): INSERT INTO wetter_daten VALUES ( 190600010 , '2023-06-01 06:00:00' , 0 , 0 ) Connection ID (thread ID): 5569194 Status: NOT_KILLED Optimizer switch: index_merge=on,index_merge_union=on,index_merge_sort_union=on,index_merge_intersection=on,index_merge_sort_intersection=off,e ngine_condition_pushdown=off,index_condition_pushdown=on,derived_merge=on,derived_with_keys=on,firstmatch=on,loosescan=on,materialization=on,in _to_exists=on,semijoin=on,partial_match_rowid_merge=on,partial_match_table_scan=on,subquery_cache=on,mrr=off,mrr_cost_based=off,mrr_sort_keys=o ff,outer_join_with_cache=on,semijoin_with_cache=on,join_cache_incremental=on,join_cache_hashed=on,join_cache_bka=on,optimize_join_buffer_size=o n,table_elimination=on,extended_keys=on,exists_to_in=on,orderby_uses_equalities=on,condition_pushdown_for_derived=on,split_materialized=on,cond ition_pushdown_for_subquery=on,rowid_filter=on,condition_pushdown_from_having=on,not_null_range_scan=off The manual page at https://mariadb.com/kb/en/how-to-produce-a-full-stack-trace-for-mysqld/ contains information that should help you find out what is causing the crash. Writing a core file... Working directory at /home/kpc/kpc.dbs Resource Limits: Limit Soft Limit Hard Limit Units Max cpu time unlimited unlimited seconds Max file size unlimited unlimited bytes Max data size unlimited unlimited bytes Max stack size 8388608 unlimited bytes Max core file size 0 unlimited bytes Max resident set unlimited unlimited bytes Max processes 127967 127967 processes Max open files 32313 32313 files Max locked memory 65536 65536 bytes Max address space unlimited unlimited bytes Max file locks unlimited unlimited locks Max pending signals 127967 127967 signals Max msgqueue size 819200 819200 bytes Max nice priority 0 0 Max realtime priority 0 0 Max realtime timeout unlimited unlimited us Core pattern: core 230601 08:22:10 mysqld_safe Number of processes running now: 0 230601 08:22:10 mysqld_safe mysqld restarted 2023-06-01 8:22:11 0 [Note] /usr/sbin/mariadbd (mysqld 10.5.15-MariaDB-0+deb11u1-log) starting as process 24005 ... 2023-06-01 8:22:11 0 [Note] mariadbd: Aria engine: starting recovery recovered pages: 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% (0.1 seconds); tables to flush: 62 61 60 59 58 57 56 55 54 53 52 51 50 49 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 (0.4 seconds); 2023-06-01 8:22:11 0 [Note] mariadbd: Aria engine: recovery done 2023-06-01 8:22:11 0 [Note] InnoDB: Uses event mutexes 2023-06-01 8:22:11 0 [Note] InnoDB: Compressed tables use zlib 1.2.11 2023-06-01 8:22:11 0 [Note] InnoDB: Number of pools: 1 2023-06-01 8:22:11 0 [Note] InnoDB: Using crc32 + pclmulqdq instructions 2023-06-01 8:22:11 0 [Note] InnoDB: Using Linux native AIO 2023-06-01 8:22:11 0 [Note] InnoDB: Initializing buffer pool, total size = 5368709120, chunk size = 134217728 2023-06-01 8:22:11 0 [Note] InnoDB: Completed initialization of buffer pool 2023-06-01 8:22:11 0 [Note] InnoDB: Starting crash recovery from checkpoint LSN=463015889782,463015889782 2023-06-01 8:22:13 0 [Note] InnoDB: Starting final batch to recover 2019 pages from redo log. 2023-06-01 8:22:15 0 [Note] InnoDB: 128 rollback segments are active. 2023-06-01 8:22:15 0 [Note] InnoDB: Removed temporary tablespace data file: "ibtmp1" 2023-06-01 8:22:15 0 [Note] InnoDB: Creating shared tablespace for temporary tables 2023-06-01 8:22:15 0 [Note] InnoDB: Setting file './ibtmp1' size to 12 MB. Physically writing the file full; Please wait ... 2023-06-01 8:22:15 0 [Note] InnoDB: File './ibtmp1' size is now 12 MB. 2023-06-01 8:22:15 0 [Note] InnoDB: 10.5.15 started; log sequence number 463033688604; transaction id 19549881 2023-06-01 8:22:15 0 [Note] InnoDB: Loading buffer pool(s) from /home/kpc/kpc.dbs/ib_buffer_pool 2023-06-01 8:22:15 0 [Note] Plugin 'FEEDBACK' is disabled. 2023-06-01 8:22:15 0 [Note] Server socket created on IP: '::'. 2023-06-01 8:22:16 0 [Note] Reading of all Master_info entries succeeded 2023-06-01 8:22:16 0 [Note] Added new Master_info '' to hash table 2023-06-01 8:22:16 0 [Note] /usr/sbin/mariadbd: ready for connections. Version: '10.5.15-MariaDB-0+deb11u1-log' socket: '/run/mysqld/mysqld.sock' port: 3306 Debian 11 2023-06-01 8:22:18 0 [Note] InnoDB: Buffer pool(s) load completed at 230601 8:22:18

2 1

[Maria-discuss] jujst re-joined trial e-mail don't read
by antonio scotti 30 May '23

30 May '23

=============================================== PermaCultura-es http://www.permacultura-es.org

1 0

[Maria-discuss] trial e-mail don't read
by antonio scotti 30 May '23

30 May '23

=============================================== PermaCultura-es http://www.permacultura-es.org

1 0

[Maria-discuss] Innodb_buffer_pool_read_requests is zero
by mariadb＠Biblestuph.com 19 May '23

19 May '23

Just upgraded from 10.6.12 to 10.6.13 on a Debian Bullseye system, and seeing this oddity. Innodb_buffer_pool_read_requests is zero, and seems to remain at zero no matter what I do (I even tried selecting all the data from a small table several times in a row with no change). It's been running a couple hours now; I've restarted it a couple times. I have 10.6.13 on a few other platforms as well (including Buster), and I'm not seeing the same issue on them. What would cause this? I have no other Bullseye systems running 10.6.13 yet, and the same system has been running 10.6.12 for over a week now without showing this issue. The system _seems_ to be running fine, it's a replica system that is being fed actively from the master (data comes in to it very regularly). MariaDB [(none)]> SHOW GLOBAL STATUS LIKE 'innodb_buffer_pool_%'; +-----------------------------------------+--------------------------------------------------+ | Variable_name | Value | +-----------------------------------------+--------------------------------------------------+ | Innodb_buffer_pool_dump_status | | | Innodb_buffer_pool_load_status | Buffer pool(s) load completed at 230511 23:21:26 | | Innodb_buffer_pool_resize_status | | | Innodb_buffer_pool_load_incomplete | OFF | | Innodb_buffer_pool_pages_data | 609103 | | Innodb_buffer_pool_bytes_data | 9979543552 | | Innodb_buffer_pool_pages_dirty | 19465 | | Innodb_buffer_pool_bytes_dirty | 318914560 | | Innodb_buffer_pool_pages_flushed | 0 | | Innodb_buffer_pool_pages_free | 299441 | | Innodb_buffer_pool_pages_made_not_young | 0 | | Innodb_buffer_pool_pages_made_young | 107 | | Innodb_buffer_pool_pages_misc | 0 | | Innodb_buffer_pool_pages_old | 224824 | | Innodb_buffer_pool_pages_total | 908544 | | Innodb_buffer_pool_pages_lru_flushed | 0 | | Innodb_buffer_pool_pages_lru_freed | 0 | | Innodb_buffer_pool_pages_split | 726 | | Innodb_buffer_pool_read_ahead_rnd | 0 | | Innodb_buffer_pool_read_ahead | 3 | | Innodb_buffer_pool_read_ahead_evicted | 0 | | Innodb_buffer_pool_read_requests | 0 | | Innodb_buffer_pool_reads | 604346 | | Innodb_buffer_pool_wait_free | 0 | | Innodb_buffer_pool_write_requests | 3639341 | +-----------------------------------------+--------------------------------------------------+ 25 rows in set (0.000 sec) MariaDB [(none)]> SHOW VARIABLES LIKE 'innodb_buffer_pool%'; +-------------------------------------+----------------+ | Variable_name | Value | +-------------------------------------+----------------+ | innodb_buffer_pool_chunk_size | 134217728 | | innodb_buffer_pool_dump_at_shutdown | ON | | innodb_buffer_pool_dump_now | OFF | | innodb_buffer_pool_dump_pct | 80 | | innodb_buffer_pool_filename | ib_buffer_pool | | innodb_buffer_pool_load_abort | OFF | | innodb_buffer_pool_load_at_startup | ON | | innodb_buffer_pool_load_now | OFF | | innodb_buffer_pool_size | 15032385536 | +-------------------------------------+----------------+ 9 rows in set (0.000 sec)

2 5

[Maria-discuss] Question regarding mariadb-dump/mysqldump and rocksdb
by Jonas Krauss 16 May '23

16 May '23

Hi MariaDB, in Myrocks / Facebook MySQL there was a useful addition to mysqldump which makes queries issued by the dumper not add to the block cache of the rocksdb instance. It was added quite some time ago: https://github.com/facebook/mysql-5.6/commit/2223479fea78a246884c5a09ef4d02… The corresponding server variable used is *rocksdb_skip_fill_cache* Is this active in MariaDB's version of mysqldump as well? The server variable exists in MariaDB's MyRocks (checked for 10.6.12) but I have not found a reference in MariaDB's source code when searching for rocksdb_skip_fill_cache. If it is not active it would be great if it could be added in a future version. Best Jonas

2 2

[Maria-discuss] Undo logs aren't truncating
by mariadb＠Biblestuph.com 15 May '23

15 May '23

On a recently updated MariaDB 10.6.13 server (Debian Bullseye) I noticed that my ibdata1 file was continuously growing; it had gotten a little over 3G after running a couple days. I would rather keep this growth in check so used it as an excuse to setup undo tablespaces that could be truncated (I had wanted to try this for some time anyway). So: Dump the whole DB, delete all the InnoDB tables, stop the server, delete the innodb system tables, reset the appropriate undo variables (with innodb_max_undo_log_size set to 750M, innodb_undo_tablespaces set to 4, and innodb_undo_log_truncate set to ON), restart the server and reload the data. I now have FOUR undo files that appear to be continuously growing; as of this writing they are over 1G and still increasing. This is an active read-only replica (processing updates very frequently; at times continuously but not constantly as such). How do I get it to actually reclaim the space? Or am I misunderstanding how this is supposed to work? Below are my current relevant settings (or at least what I think are relevant). Normally I run with innodb_purge_rseg_truncate_frequency = 128; but for testing I have it now set to 1 (but with no noticeable difference). Things I have tried (no luck on any of them): Set innodb_purge_rseg_truncate_frequency = 1. Set innodb_max_purge_lag_wait=0 (it returned nearly immediately). Stop both slave threads and wait for several minutes. Set innodb_max_purge_lag and innodb_max_purge_lag_delay (see below). MariaDB [(none)]> show variables like '%trunc%'; +--------------------------------------+-------+ | Variable_name | Value | +--------------------------------------+-------+ | innodb_purge_rseg_truncate_frequency | 1 | | innodb_undo_log_truncate | ON | +--------------------------------------+-------+ 2 rows in set (0.000 sec) MariaDB [(none)]> show variables like '%purge%'; +--------------------------------------+------------+ | Variable_name | Value | +--------------------------------------+------------+ | aria_log_purge_type | immediate | | innodb_max_purge_lag | 1000 | | innodb_max_purge_lag_delay | 5000 | | innodb_max_purge_lag_wait | 4294967295 | | innodb_purge_batch_size | 300 | | innodb_purge_rseg_truncate_frequency | 1 | | innodb_purge_threads | 4 | | relay_log_purge | ON | +--------------------------------------+------------+ 8 rows in set (0.000 sec) MariaDB [(none)]> show variables like '%undo%'; +--------------------------+-----------+ | Variable_name | Value | +--------------------------+-----------+ | innodb_max_undo_log_size | 786432000 | | innodb_undo_directory | ./ | | innodb_undo_log_truncate | ON | | innodb_undo_tablespaces | 4 | +--------------------------+-----------+ 4 rows in set (0.000 sec)

2 1

[Maria-discuss] mariadb crash und recovery
by Radisson97＠gmx.de 14 May '23

14 May '23

Hello list, i had a crash at weekend with no apparend cause. The aria engine did a recovery without furture problems. Maybe that is a hint that somebody can make use of. Hints welcome. CU from log: 230416 7:20:17 [ERROR] mysqld got signal 11 ; This could be because you hit a bug. It is also possible that this binary or one of the libraries it was linked against is corrupt, improperly built, or misconfigured. This error can also be caused by malfunctioning hardware. To report this bug, see https://mariadb.com/kb/en/reporting-bugs We will try our best to scrape up some info that will hopefully help diagnose the problem, but since we have already crashed, something is definitely wrong and this may fail. Server version: 10.5.15-MariaDB-0+deb11u1-log key_buffer_size=134217728 read_buffer_size=131072 max_used_connections=86 max_threads=252 thread_count=38 It is possible that mysqld could use up to key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 685818 K bytes of memory Hope that's ok; if not, decrease some variables in the equation. Thread pointer: 0x7f01e4000c58 Attempting backtrace. You can use the following information to find out where mysqld died. If you see no messages after this, something went terribly wrong... stack_bottom = 0x7f02b6894d38 thread_stack 0x49000 ??:0(my_print_stacktrace)[0x5559d92bb54e] ??:0(handle_fatal_signal)[0x5559d8dbaf65] sigaction.c:0(__restore_rt)[0x7f043591a140] ??:0(_ma_state_info_write)[0x5559d8fafc73] ??:0(_ma_state_info_write)[0x5559d8fb1737] ??:0(_ma_state_info_write)[0x5559d8fb946c] ??:0(handler::ha_write_row(unsigned char const*))[0x5559d8dc7aa8] ??:0(MyCTX_nopad::finish(unsigned char*, unsigned int*))[0x5559d8ff40da] ??:0(handler::ha_write_row(unsigned char const*))[0x5559d8dc7aa8] ??:0(write_record(THD*, TABLE*, st_copy_info*, select_result*))[0x5559d8b8787d] ??:0(mysql_insert(THD*, TABLE_LIST*, List<Item>&, List<List<Item> >&, List<Item>&, List<Item>&, enum_duplicates, bool, select _result*))[0x5559d8b8f8a4] ??:0(mysql_execute_command(THD*))[0x5559d8bc2716] ??:0(mysql_parse(THD*, char*, unsigned int, Parser_state*, bool, bool))[0x5559d8bc65db] ??:0(dispatch_command(enum_server_command, THD*, char*, unsigned int, bool, bool))[0x5559d8bc8a5d] ??:0(do_command(THD*))[0x5559d8bca2de] ??:0(do_handle_one_connection(CONNECT*, bool))[0x5559d8cbbfb2] ??:0(handle_one_connection)[0x5559d8cbc22d] ??:0(MyCTX_nopad::finish(unsigned char*, unsigned int*))[0x5559d8ff811b] nptl/pthread_create.c:478(start_thread)[0x7f043590eea7] x86_64/clone.S:97(__GI___clone)[0x7f0435525def] Trying to get some variables. Some pointers may be invalid and cause the dump to abort. Query (0x7f01e4010440): INSERT INTO messwerte VALUES ( 83110995 , 1200 , 909 , 331 , '2023-04-16 05:20:00' , '2023-04-16 05: 20:00' , 0.0703000000000000013766765505351941101253032684326171875 , 0.001000000000000000020816681711721685132943093776702880 859375 , 0 , 0 , 0 , 0 , 60 , 0 ) Connection ID (thread ID): 5463672 Status: NOT_KILLED Optimizer switch: index_merge=on,index_merge_union=on,index_merge_sort_union=on,index_merge_intersection=on,index_merge_sort_ intersection=off,engine_condition_pushdown=off,index_condition_pushdown=on,derived_merge=on,derived_with_keys=on,firstmatch=o n,loosescan=on,materialization=on,in_to_exists=on,semijoin=on,partial_match_rowid_merge=on,partial_match_table_scan=on,subque ry_cache=on,mrr=off,mrr_cost_based=off,mrr_sort_keys=off,outer_join_with_cache=on,semijoin_with_cache=on,join_cache_increment al=on,join_cache_hashed=on,join_cache_bka=on,optimize_join_buffer_size=on,table_elimination=on,extended_keys=on,exists_to_in= on,orderby_uses_equalities=on,condition_pushdown_for_derived=on,split_materialized=on,condition_pushdown_for_subquery=on,rowi d_filter=on,condition_pushdown_from_having=on,not_null_range_scan=off The manual page at https://mariadb.com/kb/en/how-to-produce-a-full-stack-trace-for-mysqld/ contains information that should help you find out what is causing the crash. Writing a core file... Working directory at /home/kpc/kpc.dbs Resource Limits: Limit Soft Limit Hard Limit Units Max cpu time unlimited unlimited seconds Max file size unlimited unlimited bytes Max data size unlimited unlimited bytes Max stack size 8388608 unlimited bytes Max core file size 0 unlimited bytes Max resident set unlimited unlimited bytes Max processes 127976 127976 processes Max open files 32313 32313 files Max locked memory 65536 65536 bytes Max address space unlimited unlimited bytes Max file locks unlimited unlimited locks Max pending signals 127976 127976 signals Max msgqueue size 819200 819200 bytes Max nice priority 0 0 Max realtime priority 0 0 Max realtime timeout unlimited unlimited us Core pattern: core 230416 07:20:19 mysqld_safe Number of processes running now: 0 230416 07:20:19 mysqld_safe mysqld restarted 2023-04-16 7:20:19 0 [Note] /usr/sbin/mariadbd (mysqld 10.5.15-MariaDB-0+deb11u1-log) starting as process 415 ... 2023-04-16 7:20:20 0 [Note] mariadbd: Aria engine: starting recovery recovered pages: 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% (0.0 seconds); tables to flush: 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 (0.1 seconds); 2023-04-16 7:20:20 0 [Note] mariadbd: Aria engine: recovery done

4 3

[Maria-discuss] Checkpoint Age and Redo Log Flushing in 10.5+
by Gordan Bobic 14 May '23

14 May '23

Ever since 10.5, the redo log flushing seems to have become ... suboptimal. It looks like the database does nothing at all to flush the logs until it hits the hwm, and then it goes into overdrive and flushes everything to tablespaces fast, before it goes back to doing nothing until the log fills up. This causes a problem when the redo log is already full and a large burst of writes arrives because it overwhelms the redo log and causes a stall. Is there a bug fix for this or a configuration change available to restore the old behaviour where everything gets flushed continuously and switches up a gear when the hwm is reached? Here are the screenshots comparing 10.5+ sawtooth pattern vs 10.4 and older versions.

2 1

[Maria-discuss] howto c-connect bindings
by juerg oehler 06 May '23

06 May '23

hi I'm working with|mariadb|and|c-connect|prepared statements and run into crashes executing them once a while. i so addressed my problem to stakeoverflow <https://stackoverflow.com/questions/76170354/mariadb-howto-c-connect-bindin…> hoping to get more information for binding programs data structure to db interface. i did get an answer. but this answer does not go inline with your documentation <https://mariadb.com/kb/en/mariadb-connectorc-data-structures/#mysql_bind>. i so do have the option to follow your documentation or use the answer and run into crashes in either way. you might ask why i'm not just going with your documentation? your documentation is far to thin to get a reliable idea how bind structure is used and i do get proofed this by crashes. second i played with <my_bool * is_nulll Pointer which indicates if column is NULL (not used for parameters)> for parameters and behavior was not as documented. this is my motivation to bring my problem to this mailing list hopping someone can point me to a documentation i can rely on. i found a short and useful documentation in mysql with permissible in and out types <https://dev.mysql.com/doc/c-api/8.0/en/c-api-prepared-statement-type-codes.…> and statement data structures <https://dev.mysql.com/doc/c-api/8.0/en/c-api-prepared-statement-data-struct…>. can i use these informations to build 2 test applications one focusing on parameter (attachment) one focusing on result? currently i implemented MariaPrepUpd.c using mariadb documentation. i felt like flying blind and ended up in crashes - once a while. do you think i could use mysql documentation for improving bindings? or do you have a better option? (i studied your bulk example <https://mariadb.com/kb/en/bulk-insert-row-wise-binding/>. i even was using this example as base for first try an hit the wall by using STMT_INDICATOR_NTS. i then learned my problem is not just the example using STMT_ATTR_ARRAY_SIZE = 1.) my goal is to have a version of MariaPrepUpd.c i can use for any application for iteration 1 to n. thank you for short answer. have a great day. juerg(=george) oehler

1 0

[Maria-discuss] mysql.innodb-table-stats-and mysql.innodb-index-stats are not updating properly
by ragul rangarajan 21 Apr '23

21 Apr '23

Hello all, I am facing a weird issue with my DB table where its Index and table stats not getting updated with values automatically in MariaDB 10.6.7 The table has more than a billion records. On querying the stats of the table, I could not see any stats for the table that begin automatically calculated until we do a mysql restart or ANALYZE TABLE. Once triggered we can able to see the index values are calculated and updated in this table 127.0.0.1:3307> SELECT * FROM mysql.innodb_table_stats ;+---------------+-------------------------------------+---------------------+---------+----------------------+--------------------------+| database_name | table_name | last_update | n_rows | clustered_index_size | sum_of_other_index_sizes |+---------------+-------------------------------------+---------------------+---------+----------------------+--------------------------+| DB | TableStats#P#p2023_04_11_13_00_00 | 2023-04-11 12:00:11 | 0 | 1 | 0 || DB | TableStats#P#p2023_04_11_14_00_00 | 2023-04-11 13:00:10 | 0 | 1 | 0 || DB | TableStats#P#p2023_04_11_15_00_00 | 2023-04-11 14:00:04 | 0 | 1 | 0 || DB | TableStats#P#p2023_04_11_16_00_00 | 2023-04-11 15:00:04 | 0 | 1 | 0 |+---------------+-------------------------------------+---------------------+---------+----------------------+--------------------------+ 127.0.0.1:3307> SELECT * FROM mysql.innodb_index_stats;+---------------+-------------------------------------+------------+---------------------+--------------+------------+-------------+------------------------------------------------------------+| database_name | table_name | index_name | last_update | stat_name | stat_value | sample_size | stat_description |+---------------+-------------------------------------+------------+---------------------+--------------+------------+-------------+------------------------------------------------------------+| DB | TableStats#P#p2023_04_11_16_00_00 | PRIMARY | 2023-04-11 15:00:04 | n_diff_pfx01 | 0 | 1 | AgentId || DB | TableStats#P#p2023_04_11_16_00_00 | PRIMARY | 2023-04-11 15:00:04 | n_diff_pfx02 | 0 | 1 | AgentId,rackNr || DB | TableStats#P#p2023_04_11_16_00_00 | PRIMARY | 2023-04-11 15:00:04 | n_leaf_pages | 1 | NULL | Number of leaf pages in the index || DB | TableStats#P#p2023_04_11_16_00_00 | PRIMARY | 2023-04-11 15:00:04 | size | 1 | NULL | Number of pages in the index |+---------------+-------------------------------------+------------+---------------------+--------------+------------+-------------+------------------------------------------------------------+ As per the documentation https://mariadb.com/kb/en/innodb-persistent-statistics/, enabling *innodb_stats_auto_recalc* parameter should update the index value automatically which was not happening for me even though the changes are 10% of the row got affected within few days. The same was working for other DB table which is of lower load. Is there any way to debug why the calculation was not updated properly with values or any way to reduce the sample size? *Thanks and Regards,* Ragul R

2 2

[Maria-discuss] DELETE query hanging on Statistics (10.6.12)
by mariadb＠Biblestuph.com 14 Apr '23

14 Apr '23

With this table: CREATE TABLE t1 ( label varchar(100) NOT NULL, dt datetime NOT NULL, success tinyint(1) NOT NULL, PRIMARY KEY (label,dt), KEY dt (dt), KEY success (success,dt) ) ENGINE=InnoDB DEFAULT CHARSET=latin1 ROW_FORMAT=DYNAMIC I have a DELETE query that has been stuck in Statistics for nearly an hour now: DELETE MyDB.t1 FROM MyDB.t1 WHERE dt <= '2023-02-21 08:06:46' AND dt >= '2023-02-21 08:03:48' This statement runs several times per hour to trim the table (and has for years). It has run fine against my 10.6.12 DB (this is a downstream replica DB) for a couple weeks now(when I upgraded the replica); but suddenly today produced the hang. Any ideas why? The table has nearly 15 million rows; but the rows are more or less evenly spaced across datetimes. Also I find that I cannot kill the query using KILL 12345; the KILL statement just comes back immediately without reporting an error and the statement continues. Is there no way to kill this statement without crashing the DB? Again, this is a replica statement that is executing. I had planned to stop the query, execute an ANALYZE TABLE t1 and then try it again, but I don't seem to be able to kill the query. Thank you! Dan

3 5

[Maria-discuss] Privilege Question
by Scott Canaan 06 Apr '23

06 Apr '23

We are on MariaDB 10.5.18. There is a requirement to send all syslog data to a central syslog server. In the past, we did it using a login called ITS_READ. It has limited privs on purpose, but used to be able to execute the SET GLOBAL statements that we needed. Those statements are: SET GLOBAL server_audit_output_type=SYSLOG; SET GLOBAL server_audit_logging=1; SET GLOBAL server_audit_syslog_facility=LOG_LOCAL2; SET GLOBAL server_audit_events="connect,table,query_ddl,query_dcl"; Now, it doesn't have the privs needed to do that. In looking online, I found that there is a "new" privilege (in MySQL 8), called system_variables_admin that allows the user to execute SET GLOBAL statements. However, when I try to grant that privilege, I get an error: MariaDB [(none)]> grant system_variables_admin to 'ITS_READ'@'localhost'; ERROR 1959 (OP000): Invalid role specification `system_variables_admin` It seems that the only other option is to grant the SUPER privilege, which we don't want to do. How do I get around this, or do I just tell the sysadmins they can't have the syslog data? Scott Canaan '88 Sr Database Administrator Information & Technology Services Finance & Administration Rochester Institute of Technology o: (585) 475-7886 | f: (585) 475-7520 srcdco(a)rit.edu<mailto:srcdco@rit.edu> | c: (585) 339-8659 CONFIDENTIALITY NOTE: The information transmitted, including attachments, is intended only for the person(s) or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and destroy any copies of this information.

3 3

[Maria-discuss] Disable validation for `DATETIME` columns
by William Edwards 03 Apr '23

03 Apr '23

Hi, Is it possible to disable validation for `DATETIME` columns? Context: A customer uses the Europe/Amsterdam session time zone. Therefore, they cannot insert dates during the hour in which DST starts and ends. Using the UTC session time zone would solve the issue. However, they also use auto-converting `TIMESTAMP` columns. Switching to a different session time zone would cause issues with those. Relaxing constraints with `sql_mode=NO_ZERO_DATE,NO_ZERO_IN_DATE,ALLOW_INVALID_DATES` does not suffice. With kind regards, William Edwards

2 3

[Maria-discuss] Column named 'row_number' gives error in 10.11
by Keith Edmunds 02 Apr '23

02 Apr '23

I have a table that includes an int column named row_number. This worked fine until I upgraded Debian from 11 to 12 which upgraded MariaDB from 10.5 to 10.11. I now get errors when referencing that column for updates. If I put backticks around the column name the update runs without error. Is this an expected behaviour change from 10.5 → 10.11? Example: CREATE DATABASE test; USE test; CREATE TABLE `kae` ( `id` int(11) NOT NULL, `row_number` int(11) NOT NULL ) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_general_ci; INSERT INTO `kae` (`id`, `row_number`) VALUES (1, 0); UPDATE kae SET row_number=12 WHERE id=1; ERROR 1064 (42000): You have an error in your SQL syntax; check the manual that corresponds to your MariaDB server version for the right syntax to use near 'row_number=12 WHERE id=1' at line 1

4 8

[Maria-discuss] Mounting tmpdir via NFS
by Simon Avery 29 Mar '23

29 Mar '23

Hello, I have a question that I've been struggling to get a clear answer to, or even any indication someone has done this before, and would appreciate any experience or thoughts. Are there any known issues with defining tmpdir on multiple servers to point to a single NFS-mounted network share? Would each instance co-habit peacefully and not try to change any files from another instance? Why do I want to do this? I have around 80 Centos/Rocky vms which run Mariadb 10.6.x. Due to the nature of the work they do, some of them require a significant amount of tmpdir (up to 75Gb) relatively briefly (once or twice a day for a few minutes) This growth is not always predictable, so generally the first I know about it is when Mariadb reports lowspace and the query stops. We then grow the /tmp partition that is defined as tmpdir and if we're quick enough, the query completes, but often the job dies. Then that extra space stays locked away forever, even if that server never tries to use that much tmpdir again. This does not feel efficient and I am pondering better ways of doing this. One such thought would be to have a single HA NFS share on our network, mounted onto each VM as something like /mariadb-tempdir, and set $tmpdir to it. In my mind, this would make a better use of one large allocation of space used by many things often, rather than lots of individual small to large sized bits of space used by one thing, rarely. The chance of two or more servers filling that space feels small, but if it happened, one would wait until space was cleared - which might actually happen in this shared scenario when it never does in a single-server situation. In my particular world, we have fast storage and a fast network. Speed tests on large and tiny files over NFS give surprisingly similar performance to writing to local vm disks, so I'm not worried about the overhead of NFS and even if it was, slow might still be preferred to full. Thank you, Simon

3 5

[Maria-discuss] MariaDB Online-DDL
by Rhys.Campbell＠swisscom.com 17 Mar '23

17 Mar '23

Hello All, I'm back working with MariaDB again after a break of some years. While InnoDB Online-DDL had appeared when I was working with it, it was still sometimes a challenge to manage schema changes for very large tables. https://mariadb.com/kb/en/innodb-online-ddl-overview/ Can anyone summarize how InnoDB stacks up for Online-DDL now? Is it comparable to MSSQL (which I found to be pretty good for tables in the many tens of GB)? Are workarounds like pt-online-schema-change and blue/green deployments still required? Are there any specific areas to watch out for? Any feedback appreciated. Cheers, Rhys

2 1

[Maria-discuss] can not kill process
by Radisson97＠gmx.de 17 Mar '23

17 Mar '23

hi i have a strange effect a simple insert got stuck (why ?) i tried to kill but no success, it is even marked as "killed" (see below) 1398867 kpc localhost kpc 1716 Killed INSERT INTO i can kill other processes- my system: 10.5.15-MariaDB engine aria 5.10.0-14-amd64 any ideas what may the problem ? (no usefull entries in any log, as you expected ) re,

3 4

[Maria-discuss] Optimized way to build MariaDB
by Otto Kekäläinen 12 Mar '23

12 Mar '23

Hi! I have spent a lot of time teaching various people how to build (and re-build) MariaDB. If you are interested to learn how I use entr + cmake --fresh + ninja + clang + ccache check out https://optimizedbyotto.com/post/quick-builds-and-rebuilds-of-mariadb-with-… - Otto

1 0

[Maria-discuss] MariaDB Connector C 3.1 ABI contract broken in latest release
by Otto Kekäläinen 04 Mar '23

04 Mar '23

Hello! After uploading MariaDB 10.3.38 in Debian I started seeing reports about libmariadb3 being broken for some consumers. Turns out that the function mysql_get_client_version() had a major behavior change in the latest MariaDB Connector C 3.1.20 which is embedded in the MariaDB 10.3 build. Details in PR https://github.com/mariadb-corporation/mariadb-connector-c/pull/219. Affected library consumers are listed in https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1031863#86 Unfortunately it might be too late to fix this, the milk is already spilled. Could you please in future keep as the engineering policy that *libraries are not allowed to have backwards-incompatible changes* unless the soname/version is also bumped. You might add a new symbol in an existing library version without causing breakage in library consumers, but you cannot remove any symbols/functions, or change their behavior significantly. In this case you completely changed what version the library returns. Please acknowledge these base rules and be more diligent in the future to avoid risks. Also, if you want to introduce libmariadb4, please do it on MariaDB 11.0 or newer only, and don't change any ABI in existing stable releases of libmariadb3. Thanks!

1 0

[Maria-discuss] EXPLAIN Analyser / Usage statistics
by Colin Charles 02 Mar '23

02 Mar '23

Hi, What might have happened to the EXPLAIN Analyzer https://mariadb.org/explain_analyzer/ It is now returning 404 not found Similarly, if you click https://mariadb.org/about/#usage-statistics (usage statistics at the bottom of every page) you’re also getting things like “oops that page can’t be found” e.g. https://mariadb.org/feedback_plugin/stats/server_count_by_month/ While I’m at it, it might also be that the download stats haven’t been updated since January? https://mariadb.org/statistics/ (I think Andrew fixed this in December, but maybe the Docker container died again) cheers, -c -- Colin Charles, http://bytebot.net/blog/ twitter: @bytebot | skype: colincharles "First they ignore you, then they laugh at you, then they fight you, then you win." -- Mohandas Gandhi

2 2

[Maria-discuss] Creating a trigger on information_schema.processlist
by Gordan Bobic 28 Feb '23

28 Feb '23

A long standing annoyance has been that stored procedures are opaque when it comes to profiling them with more granularity than just the entire procedure. SHOW FULL PROCESSLIST shows the individual queries while they are running, but polling SHOW FULL PROCESSLIST or SELECT * FROM information_schema.PROCESSLIST is not a practical solution because you can only poll it so many times per second without crippling the server. So I had the idea to hook triggers to information_schema.PROCESSLIST to capture the changes to the INFO column and shunt them into mysql.slow_log. But MariaDB informs them that root isn't privileged enough to crate a trigger in information_schema. Is there another way to augment slow logging granularity of queries that execute as part of the stored procedure?

2 1

[Maria-discuss] Mariadb 10.4 to 10.6 performance regression
by Reinis Rozitis 30 Jan '23

30 Jan '23

Hello, for testing the upgrade path I did an in-place slave server (with optimistic parallel replication) upgrade from 10.4.26 to 10.6.11, but the performance difference is so significant/abysmal I don't even know where to look at (see the attached QPS chart before and after upgrade - it's like 2-3x times lower!). All the previous settings/configuration are kept. I have various server parameters monitored but none seem to indicate why would everything become way slower (see for example the disk write operations also are 2-3 times lower). Maybe someone can give an idea where to look or what to check what could make such an impact? p.s. I know about https://jira.mariadb.org/browse/MDEV-29988 but I don't think that I'm hitting it since the cpu load hasn't changed (is even lower). wbr rr

1 1

[Maria-discuss] Branch 10.4 conflict
by Otto Kekäläinen 30 Jan '23

30 Jan '23

Hello! My local branch of 10.4 was at commit https://github.com/MariaDB/server/commit/9335e2a2b1d555d2ad83ae6f44acb8467c… and pulling from upstream causes now conflicts. Did somebody (Jan?) overwrite git history recently for mariadb-server or what might be going on? - Otto

3 2

[Maria-discuss] base64 text
by Ruben Safir 29 Jan '23

29 Jan '23

is there a way to encode text blogs to base64 so I can handle emails that are stored in the database so they don't look like binary trash like this? http://www.nylxs.com/messages.html?id=553339&archive=2023-01-01 ------ So many immigrant groups have swept through our town that Brooklyn, like Atlantis, reaches mythological proportions in the mind of the world - RI Safir 1998 http://www.mrbrklyn.com DRM is THEFT - We are the STAKEHOLDERS - RI Safir 2002 http://www.nylxs.com - Leadership Development in Free Software http://www.brooklyn-living.com Being so tracked is for FARM ANIMALS and extermination camps, but incompatible with living as a free human being. -RI Safir 2013

2 1

[Maria-discuss] subscribe
by Ruben Safir 29 Jan '23

29 Jan '23

subscribe -- So many immigrant groups have swept through our town that Brooklyn, like Atlantis, reaches mythological proportions in the mind of the world - RI Safir 1998 http://www.mrbrklyn.com DRM is THEFT - We are the STAKEHOLDERS - RI Safir 2002 http://www.nylxs.com - Leadership Development in Free Software http://www.brooklyn-living.com Being so tracked is for FARM ANIMALS and extermination camps, but incompatible with living as a free human being. -RI Safir 2013

1 0