Re: [Maria-discuss] MEMORY tables and binlog
Dajka Tamás writes: > Hi All, > > I’m facing a weird error. We’re migrating our old 10.1 multi-master setup to > a new 10.3 cluster > (everything is running on Debian). New-node01 is connected as a slave to the > old cluster’s master. > > When I tried to set up the replication between the 2 new nodes (with data > files copy), the slave > thread stopped nearly immediatelly stating, that the command is invalid – > containing illegal > characters. After some investigation, it turned out, that binlog on the > master (new-node01) is – > let’s say – corrupted. After a few iterations/checking I saw, that the error > always happens with > some MEMORY tables (I know I can exclude them from the replica, but want to > understand the issue > first). > > new-node01/02 running 10.3.27-MariaDB-0+deb10u1-log on latest Debian 10.9 > (replication from > old-node is OK), replication is GTID based. HW is a new HP Gen10 DL380. > > When checking the binlog on the new-node01 with mysqlbinlog, I get the > following: > > # at 2361 > > #210407 11:13:34 server id 4 end_log_pos 2403 CRC32 0x90080d27 GTID > 100-4-37336878143 > trans > > /*!11 SET @@session.gtid_seq_no=37336878143*//*!*/; > > START TRANSACTION > > /*!*/; > > # at 2403 > > #210407 11:13:34 server id 4 end_log_pos 2527 CRC32 0x27183d2f Query > thread_id= > 341297541 exec_time=318 error_code=0 > > SET TIMESTAMP=1617786814/*!*/; > > DELETE FROM `mydb1`.`collect_output`e#<98> Looks very much related to binlog checksum. Cannot say anthing more, maybe you could develop it further knowing details that unreported here? Cheers, Andrei > > /*!*/; > > # at 2527 > > #210407 11:13:34 server id 4 end_log_pos 2598 CRC32 0x427f0ed7 > Table_map: `mydb1`. > `collect_output` mapped to number 1330 > > Please notice the illegal chars at the end of the „DELETE FROM” line. ___ Mailing list: https://launchpad.net/~maria-discuss Post to : maria-discuss@lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-discuss More help : https://help.launchpad.net/ListHelp
Re: [Maria-discuss] MEMORY tables and binlog
Issue created as MDEV-25358. > -Original Message- > From: Maria-discuss [mailto:maria-discuss- > bounces+viper=vipernet...@lists.launchpad.net] On Behalf Of Dajka Tamás > Sent: Wednesday, April 7, 2021 3:18 PM > To: 'Sergei Golubchik' > Cc: maria-discuss@lists.launchpad.net > Subject: Re: [Maria-discuss] MEMORY tables and binlog > > Hi Sergei! > > Sure, I'll report it. > Yes, I think it's correctly reproducable. > > Cheers, > > Tom > > > -Original Message- > > From: Sergei Golubchik [mailto:s...@mariadb.org] > > Sent: Wednesday, April 7, 2021 3:11 PM > > To: Dajka Tamás > > Cc: maria-discuss@lists.launchpad.net > > Subject: Re: [Maria-discuss] MEMORY tables and binlog > > > > Hi, Dajka! > > > > This looks like a bug. > > Could you report it at jira.mariadb.org? > > Do you know when it happen, can you reproduce it at will? > > > > On Apr 07, Dajka Tamás wrote: > > > Hi All, > > > > > > When checking the binlog on the new-node01 with mysqlbinlog, I get > > > the > > > following: > > > > > > # at 2361 > > > #210407 11:13:34 server id 4 end_log_pos 2403 CRC32 0x90080d27 GTID > > > 100-4-37336878143 trans > > > /*!11 SET @@session.gtid_seq_no=37336878143*//*!*/; > > > START TRANSACTION > > > /*!*/; > > > # at 2403 > > > #210407 11:13:34 server id 4 end_log_pos 2527 CRC32 0x27183d2f Query > > > thread_id=341297541 exec_time=318 error_code=0 SET > > > TIMESTAMP=1617786814/*!*/; DELETE FROM > > > `mydb1`.`collect_output`e#<98> > > > /*!*/; > > > # at 2527 > > > #210407 11:13:34 server id 4 end_log_pos 2598 CRC32 0x427f0ed7 > > > Table_map: `mydb1`.`collect_output` mapped to number 1330 > > > > > > Is this a bug or a known issue? Is there any other way to solve > > > this, than exclude the MEMORY tables from the replication? > > > > Regards, > > Sergei > > VP of MariaDB Server Engineering > > and secur...@mariadb.org > > > ___ > Mailing list: https://launchpad.net/~maria-discuss > Post to : maria-discuss@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@lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-discuss More help : https://help.launchpad.net/ListHelp
Re: [Maria-discuss] MEMORY tables and binlog
Hi Sergei! Sure, I'll report it. Yes, I think it's correctly reproducable. Cheers, Tom > -Original Message- > From: Sergei Golubchik [mailto:s...@mariadb.org] > Sent: Wednesday, April 7, 2021 3:11 PM > To: Dajka Tamás > Cc: maria-discuss@lists.launchpad.net > Subject: Re: [Maria-discuss] MEMORY tables and binlog > > Hi, Dajka! > > This looks like a bug. > Could you report it at jira.mariadb.org? > Do you know when it happen, can you reproduce it at will? > > On Apr 07, Dajka Tamás wrote: > > Hi All, > > > > When checking the binlog on the new-node01 with mysqlbinlog, I get the > > following: > > > > # at 2361 > > #210407 11:13:34 server id 4 end_log_pos 2403 CRC32 0x90080d27 GTID > > 100-4-37336878143 trans > > /*!11 SET @@session.gtid_seq_no=37336878143*//*!*/; > > START TRANSACTION > > /*!*/; > > # at 2403 > > #210407 11:13:34 server id 4 end_log_pos 2527 CRC32 0x27183d2f Query > > thread_id=341297541 exec_time=318 error_code=0 SET > > TIMESTAMP=1617786814/*!*/; DELETE FROM > > `mydb1`.`collect_output`e#<98> > > /*!*/; > > # at 2527 > > #210407 11:13:34 server id 4 end_log_pos 2598 CRC32 0x427f0ed7 > > Table_map: `mydb1`.`collect_output` mapped to number 1330 > > > > Is this a bug or a known issue? Is there any other way to solve this, > > than exclude the MEMORY tables from the replication? > > Regards, > Sergei > VP of MariaDB Server Engineering > and secur...@mariadb.org ___ Mailing list: https://launchpad.net/~maria-discuss Post to : maria-discuss@lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-discuss More help : https://help.launchpad.net/ListHelp
Re: [Maria-discuss] MEMORY tables and binlog
Hi, Dajka! This looks like a bug. Could you report it at jira.mariadb.org? Do you know when it happen, can you reproduce it at will? On Apr 07, Dajka Tamás wrote: > Hi All, > > When checking the binlog on the new-node01 with mysqlbinlog, I get the > following: > > # at 2361 > #210407 11:13:34 server id 4 end_log_pos 2403 CRC32 0x90080d27 GTID > 100-4-37336878143 trans > /*!11 SET @@session.gtid_seq_no=37336878143*//*!*/; > START TRANSACTION > /*!*/; > # at 2403 > #210407 11:13:34 server id 4 end_log_pos 2527 CRC32 0x27183d2f Query > thread_id=341297541 exec_time=318 error_code=0 > SET TIMESTAMP=1617786814/*!*/; > DELETE FROM `mydb1`.`collect_output`e#<98> > /*!*/; > # at 2527 > #210407 11:13:34 server id 4 end_log_pos 2598 CRC32 0x427f0ed7 Table_map: > `mydb1`.`collect_output` mapped to number 1330 > > Is this a bug or a known issue? Is there any other way to solve this, than > exclude the MEMORY tables from the replication? Regards, Sergei VP of MariaDB Server Engineering and secur...@mariadb.org ___ Mailing list: https://launchpad.net/~maria-discuss Post to : maria-discuss@lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-discuss More help : https://help.launchpad.net/ListHelp
Re: [Maria-discuss] MEMORY tables and binlog
Am 07.04.21 um 14:51 schrieb Antony Stone: On Wednesday 07 April 2021 at 14:48:55, Reindl Harald wrote: Am 07.04.21 um 14:46 schrieb Dajka Tamás: Hi All, I’m facing a weird error. We’re migrating our old 10.1 multi-master setup to a new 10.3 cluster (everything is running on Debian). New-node01 is connected as a slave to the old cluster’s master. When I tried to set up the replication between the 2 new nodes (with data files copy), the slave thread stopped nearly immediatelly stating, that the command is invalid – containing illegal characters. After some investigation, it turned out, that binlog on the master (new-node01) is – let’s say – corrupted. After a few iterations/checking I saw, that the error always happens with some MEMORY tables (I know I can exclude them from the replica, but want to understand the issue first) wow - 10 years after stop using memory tables at all because again and again broken replication they are still a problem? Does that explain illegal characters such as DELETE FROM `mydb1`.`collect_output`e#<98> nope - in our case it was random "record don't exist" errors when replicate a delete-statement i finally had enough and converted them all to MyISAM for the sake of sanity ___ Mailing list: https://launchpad.net/~maria-discuss Post to : maria-discuss@lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-discuss More help : https://help.launchpad.net/ListHelp
Re: [Maria-discuss] MEMORY tables and binlog
On Wednesday 07 April 2021 at 14:48:55, Reindl Harald wrote: > Am 07.04.21 um 14:46 schrieb Dajka Tamás: > > Hi All, > > > > I’m facing a weird error. We’re migrating our old 10.1 multi-master > > setup to a new 10.3 cluster (everything is running on Debian). > > New-node01 is connected as a slave to the old cluster’s master. > > > > When I tried to set up the replication between the 2 new nodes (with > > data files copy), the slave thread stopped nearly immediatelly stating, > > that the command is invalid – containing illegal characters. After some > > investigation, it turned out, that binlog on the master (new-node01) is > > – let’s say – corrupted. After a few iterations/checking I saw, that the > > error always happens with some MEMORY tables (I know I can exclude them > > from the replica, but want to understand the issue first) > > wow - 10 years after stop using memory tables at all because again and > again broken replication they are still a problem? Does that explain illegal characters such as > > DELETE FROM `mydb1`.`collect_output`e#<98> ? Antony. -- Tinned food was developed for the British Navy in 1813. The tin opener was not invented until 1858. Please reply to the list; please *don't* CC me. ___ Mailing list: https://launchpad.net/~maria-discuss Post to : maria-discuss@lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-discuss More help : https://help.launchpad.net/ListHelp
Re: [Maria-discuss] MEMORY tables and binlog
Am 07.04.21 um 14:46 schrieb Dajka Tamás: Hi All, I’m facing a weird error. We’re migrating our old 10.1 multi-master setup to a new 10.3 cluster (everything is running on Debian). New-node01 is connected as a slave to the old cluster’s master. When I tried to set up the replication between the 2 new nodes (with data files copy), the slave thread stopped nearly immediatelly stating, that the command is invalid – containing illegal characters. After some investigation, it turned out, that binlog on the master (new-node01) is – let’s say – corrupted. After a few iterations/checking I saw, that the error always happens with some MEMORY tables (I know I can exclude them from the replica, but want to understand the issue first) wow - 10 years after stop using memory tables at all because again and again broken replication they are still a problem? ___ Mailing list: https://launchpad.net/~maria-discuss Post to : maria-discuss@lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-discuss More help : https://help.launchpad.net/ListHelp
[Maria-discuss] MEMORY tables and binlog
Hi All, I'm facing a weird error. We're migrating our old 10.1 multi-master setup to a new 10.3 cluster (everything is running on Debian). New-node01 is connected as a slave to the old cluster's master. When I tried to set up the replication between the 2 new nodes (with data files copy), the slave thread stopped nearly immediatelly stating, that the command is invalid - containing illegal characters. After some investigation, it turned out, that binlog on the master (new-node01) is - let's say - corrupted. After a few iterations/checking I saw, that the error always happens with some MEMORY tables (I know I can exclude them from the replica, but want to understand the issue first). new-node01/02 running 10.3.27-MariaDB-0+deb10u1-log on latest Debian 10.9 (replication from old-node is OK), replication is GTID based. HW is a new HP Gen10 DL380. When checking the binlog on the new-node01 with mysqlbinlog, I get the following: # at 2361 #210407 11:13:34 server id 4 end_log_pos 2403 CRC32 0x90080d27 GTID 100-4-37336878143 trans /*!11 SET @@session.gtid_seq_no=37336878143*//*!*/; START TRANSACTION /*!*/; # at 2403 #210407 11:13:34 server id 4 end_log_pos 2527 CRC32 0x27183d2f Query thread_id=341297541 exec_time=318 error_code=0 SET TIMESTAMP=1617786814/*!*/; DELETE FROM `mydb1`.`collect_output`e#<98> /*!*/; # at 2527 #210407 11:13:34 server id 4 end_log_pos 2598 CRC32 0x427f0ed7 Table_map: `mydb1`.`collect_output` mapped to number 1330 Please notice the illegal chars at the end of the "DELETE FROM" line. (DELETE FROM is normal, since new-node01 is starting up from shutdown and mydb1.collect_output is a MEMORY table -> will delete the contents, so that the replications slaves have the same data). Is this a bug or a known issue? Is there any other way to solve this, than exclude the MEMORY tables from the replication? Thanks, Tom ___ Mailing list: https://launchpad.net/~maria-discuss Post to : maria-discuss@lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-discuss More help : https://help.launchpad.net/ListHelp
