Re: Replication suddenly stops on mysql 4.1.7 with Slave_IO_Running: No
We've had very good performance with the official mysql-icc-binaries, so I upgraded to 4.1.8 last weekend since there is no official 4.1.9 binary on the mysql.com-site... It didn't help with my problems, I still have replication-crashs almost every other hour. I put a fresh snapshot from the master onto the slave but it didn't help either :( A simple slave start helps, so I have a cronjob running right now checking for the replication-status and issuing a slave start if necessary I have no other idea but try the gcc-4.1.9 in about 3 weeks, I have no possibility to take the master database down anytime before that :( Gleb Paharenko schrieb: Hello. But I use 4.1.7, not 4.0.21 ...weird. As said at: http://dev.mysql.com/doc/mysql/en/news-4-1-8.html Fixed a bug which caused a crash when only the slave I/O thread was stopped and started. (Bug #6148) I suggest you to upgrade to the latest release (4.1.9 now). Jan Kirchhoff [EMAIL PROTECTED] wrote: Gleb Paharenko schrieb: Hello. I've looked through the bug database, and the only thing that I've found was an already-closed bug: http://bugs.mysql.com/bug.php?id=6148 I had been looking around the Changelogs, but I had not found that one. Sounds pretty much like my problem :( But I use 4.1.7, not 4.0.21 ...weird. Check that your server passes rpl_relayspace.test. Go to the mysql-test directory and execute: ./mysql-test-run t/rpl_relayspace.test This one runs wirhout errors on the master and the slave...: hostname:/usr/local/mysql-standard-4.1.7-pc-linux-i686-icc-glibc23/mysql-test# ./mysql-test-run t/rpl_relayspace.test Installing Test Databases Removing Stale Files Installing Master Databases running ../bin/mysqld --no-defaults --bootstrap --skip-grant-tables --basedir=.. --datadir=mysql-test/var/master-data --skip-innodb --skip-ndbcluster --skip-bdb Installing Slave Databases running ../bin/mysqld --no-defaults --bootstrap --skip-grant-tables --basedir=.. --datadir=mysql-test/var/slave-data --skip-innodb --skip-ndbcluster --skip-bdb Manager disabled, skipping manager start. Loading Standard Test Databases Starting Tests TESTRESULT --- rpl_relayspace [ pass ] --- Ending Tests Shutting-down MySQL daemon Master shutdown finished Slave shutdown finished All 1 tests were successful. I'm not able to exchange the mysql-software itself (I use the icc-binary) to a gcc-version or to upgrade to 4.1.9 in the next 2-3 weeks. And looking at the changelogs on mysql.com I don't think it would change anything... Hasn't anybody else had such problems with 4.1.x? hostname:/usr/local/mysql-standard-4.1.7-pc-linux-i686-icc-glibc23/bin# ./mysqld --version ./mysqld Ver 4.1.7-standard for pc-linux on i686 (Official MySQL-standard binary) (more detailed information on my systems in my initial mail from 2005-1-27) btw: I also ran mysqlcheck -q and mysqlcheck -o on all tables last week to make sure the tables are OK... Jan Kirchhoff [EMAIL PROTECTED] wrote: Hi, My problem still goes on... After having had the problem 2 more times within 1 day, I decided to re-do the replication (copy the whole database onto the slave with rsync and reset master and slave). That only lasted for little more than 1 day and I ended up with the same error: Could not parse relay log event entry. The possible reasons are: the master's binary log is corrupted (you can check this by running 'mysqlbinlog' on the binary log), the slave's relay log is corrupted (you can check this by running 'mysqlbinlog' on the relay log), a network problem, or a bug in the master's or slave's MySQL code. If you want to check the master's binary log or slave's relay log, you will be able to know their names by issuing 'SHOW SLAVE STATUS' on this slave.
Re: Replication suddenly stops on mysql 4.1.7 with Slave_IO_Running: No
Gleb Paharenko schrieb: Hello. I've looked through the bug database, and the only thing that I've found was an already-closed bug: http://bugs.mysql.com/bug.php?id=6148 I had been looking around the Changelogs, but I had not found that one. Sounds pretty much like my problem :( But I use 4.1.7, not 4.0.21 ...weird. Check that your server passes rpl_relayspace.test. Go to the mysql-test directory and execute: ./mysql-test-run t/rpl_relayspace.test This one runs wirhout errors on the master and the slave...: hostname:/usr/local/mysql-standard-4.1.7-pc-linux-i686-icc-glibc23/mysql-test# ./mysql-test-run t/rpl_relayspace.test Installing Test Databases Removing Stale Files Installing Master Databases running ../bin/mysqld --no-defaults --bootstrap --skip-grant-tables --basedir=.. --datadir=mysql-test/var/master-data --skip-innodb --skip-ndbcluster --skip-bdb Installing Slave Databases running ../bin/mysqld --no-defaults --bootstrap --skip-grant-tables --basedir=.. --datadir=mysql-test/var/slave-data --skip-innodb --skip-ndbcluster --skip-bdb Manager disabled, skipping manager start. Loading Standard Test Databases Starting Tests TESTRESULT --- rpl_relayspace [ pass ] --- Ending Tests Shutting-down MySQL daemon Master shutdown finished Slave shutdown finished All 1 tests were successful. I'm not able to exchange the mysql-software itself (I use the icc-binary) to a gcc-version or to upgrade to 4.1.9 in the next 2-3 weeks. And looking at the changelogs on mysql.com I don't think it would change anything... Hasn't anybody else had such problems with 4.1.x? hostname:/usr/local/mysql-standard-4.1.7-pc-linux-i686-icc-glibc23/bin# ./mysqld --version ./mysqld Ver 4.1.7-standard for pc-linux on i686 (Official MySQL-standard binary) (more detailed information on my systems in my initial mail from 2005-1-27) btw: I also ran mysqlcheck -q and mysqlcheck -o on all tables last week to make sure the tables are OK... Jan Kirchhoff [EMAIL PROTECTED] wrote: Hi, My problem still goes on... After having had the problem 2 more times within 1 day, I decided to re-do the replication (copy the whole database onto the slave with rsync and reset master and slave). That only lasted for little more than 1 day and I ended up with the same error: Could not parse relay log event entry. The possible reasons are: the master's binary log is corrupted (you can check this by running 'mysqlbinlog' on the binary log), the slave's relay log is corrupted (you can check this by running 'mysqlbinlog' on the relay log), a network problem, or a bug in the master's or slave's MySQL code. If you want to check the master's binary log or slave's relay log, you will be able to know their names by issuing 'SHOW SLAVE STATUS' on this slave. I can look at the binlog with mysqlbinlog on the master and the slave; no errors or problems. After a simple SLAVE START without having done any changes to the database, the slave thread startet again and caught up with the master. I've been using mysql's replication-feature since it first came up in 1999 or 2000 and dealt with lots of problems and workarounds, but this one is weird. Any ideas anybody? Jan -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
Re: Replication suddenly stops on mysql 4.1.7 with Slave_IO_Running: No
Hello. But I use 4.1.7, not 4.0.21 ...weird. As said at: http://dev.mysql.com/doc/mysql/en/news-4-1-8.html Fixed a bug which caused a crash when only the slave I/O thread was stopped and started. (Bug #6148) I suggest you to upgrade to the latest release (4.1.9 now). Jan Kirchhoff [EMAIL PROTECTED] wrote: Gleb Paharenko schrieb: Hello. I've looked through the bug database, and the only thing that I've found was an already-closed bug: http://bugs.mysql.com/bug.php?id=6148 I had been looking around the Changelogs, but I had not found that one. Sounds pretty much like my problem :( But I use 4.1.7, not 4.0.21 ...weird. Check that your server passes rpl_relayspace.test. Go to the mysql-test directory and execute: ./mysql-test-run t/rpl_relayspace.test This one runs wirhout errors on the master and the slave...: hostname:/usr/local/mysql-standard-4.1.7-pc-linux-i686-icc-glibc23/mysql-test# ./mysql-test-run t/rpl_relayspace.test Installing Test Databases Removing Stale Files Installing Master Databases running ../bin/mysqld --no-defaults --bootstrap --skip-grant-tables --basedir=.. --datadir=mysql-test/var/master-data --skip-innodb --skip-ndbcluster --skip-bdb Installing Slave Databases running ../bin/mysqld --no-defaults --bootstrap --skip-grant-tables --basedir=.. --datadir=mysql-test/var/slave-data --skip-innodb --skip-ndbcluster --skip-bdb Manager disabled, skipping manager start. Loading Standard Test Databases Starting Tests TESTRESULT --- rpl_relayspace [ pass ] --- Ending Tests Shutting-down MySQL daemon Master shutdown finished Slave shutdown finished All 1 tests were successful. I'm not able to exchange the mysql-software itself (I use the icc-binary) to a gcc-version or to upgrade to 4.1.9 in the next 2-3 weeks. And looking at the changelogs on mysql.com I don't think it would change anything... Hasn't anybody else had such problems with 4.1.x? hostname:/usr/local/mysql-standard-4.1.7-pc-linux-i686-icc-glibc23/bin# ./mysqld --version ./mysqld Ver 4.1.7-standard for pc-linux on i686 (Official MySQL-standard binary) (more detailed information on my systems in my initial mail from 2005-1-27) btw: I also ran mysqlcheck -q and mysqlcheck -o on all tables last week to make sure the tables are OK... Jan Kirchhoff [EMAIL PROTECTED] wrote: Hi, My problem still goes on... After having had the problem 2 more times within 1 day, I decided to re-do the replication (copy the whole database onto the slave with rsync and reset master and slave). That only lasted for little more than 1 day and I ended up with the same error: Could not parse relay log event entry. The possible reasons are: the master's binary log is corrupted (you can check this by running 'mysqlbinlog' on the binary log), the slave's relay log is corrupted (you can check this by running 'mysqlbinlog' on the relay log), a network problem, or a bug in the master's or slave's MySQL code. If you want to check the master's binary log or slave's relay log, you will be able to know their names by issuing 'SHOW SLAVE STATUS' on this slave. I can look at the binlog with mysqlbinlog on the master and the slave; no errors or problems. After a simple SLAVE START without having done any changes to the database, the slave thread startet again and caught up with the master. I've been using mysql's replication-feature since it first came up in 1999 or 2000 and dealt with lots of problems and workarounds, but this one is weird. Any ideas anybody? Jan -- For technical support contracts, goto https://order.mysql.com/?ref=ensita This email is sponsored by Ensita.NET http://www.ensita.net/ __ ___ ___ __ / |/ /_ __/ __/ __ \/ /Gleb Paharenko / /|_/ / // /\ \/ /_/ / /__ [EMAIL PROTECTED] /_/ /_/\_, /___/\___\_\___/ MySQL AB / Ensita.NET ___/ www.mysql.com -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
Re: Replication suddenly stops on mysql 4.1.7 with Slave_IO_Running: No
Hi, My problem still goes on... After having had the problem 2 more times within 1 day, I decided to re-do the replication (copy the whole database onto the slave with rsync and reset master and slave). That only lasted for little more than 1 day and I ended up with the same error: Could not parse relay log event entry. The possible reasons are: the master's binary log is corrupted (you can check this by running 'mysqlbinlog' on the binary log), the slave's relay log is corrupted (you can check this by running 'mysqlbinlog' on the relay log), a network problem, or a bug in the master's or slave's MySQL code. If you want to check the master's binary log or slave's relay log, you will be able to know their names by issuing 'SHOW SLAVE STATUS' on this slave. I can look at the binlog with mysqlbinlog on the master and the slave; no errors or problems. After a simple SLAVE START without having done any changes to the database, the slave thread startet again and caught up with the master. I've been using mysql's replication-feature since it first came up in 1999 or 2000 and dealt with lots of problems and workarounds, but this one is weird. Any ideas anybody? Jan Hello, I have a replication setup on to linux boxes (debian woody, kernel 2.4.21-xfs, mysql 4.1.7-standard official intel-compiler binary from mysql.com). master:~# mysqladmin status Uptime: 464848 Threads: 10 Questions: 296385136 Slow queries: 1752 Opens: 2629 Flush tables: 1 Open tables: 405 Queries per second avg: 637.596 slave:~# mysqladmin status Uptime: 463460 Threads: 2 Questions: 292885156 Slow queries: 6 Opens: 2510 Flush tables: 1 Open tables: 327 Queries per second avg: 631.953 both systems have identical hardware (P4 2.4ghz, 3GB RAM, SCSI-Hardware-RAID) connection is gigabit-ethernet. Everything used to work fine, but I wanted to get rid of InnoDB since I did only use that for very big table containing historical data and those tables were moved to another server. I ran out of discspace, innodb-datafiles can only grow but not shrink and i didn't need it anyway, so it had to go. I stopped the slave, changed all left over innodb-tables to myisam, added skip-innodb to my.cnf on the master and the slave, restarted the server, renewed the replication by doing it the classical way: flush tables with read log, copy the /var/lib/mysql on the slave (not much, just around 20GB), reset master, unlock tables. Then start the slave-mysqld, reset slave, slave start. Everything was fine and very fast for 4 days (from saturday till wednesday afternoon), then suddenly the slave stopped. this is where the weird stuff starts: show slave status tells me everything is fine, just Slave_IO_Running: No is wrong. After typing slave start, it says Slave_IO_Running: Yes, and Slave_SQL_Running: No. Very strange. Now i did a slave stop;slave start; and everything is fine again, the slave catches up and goes on. Today (thursday afternoon), the same thing happens again and can be solved again by slave stop;slave start;. Now it happened again around 10pm. Again, the stop-start-trick made it working again. I add the output of my mysql-shell Can anybody help me with that? This is a production system under heavy load and I can't play around with different mysql-versions and such... If I don't find a solution really quick, I'll have to do help myself with some shell-skript-daemon checking if replication is running and issuing stop slave;start slave-commands otherwise... not really the way it should be :( Thanks Jan SLAVE: slave:~# cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 15 model : 2 model name : Intel(R) Pentium(R) 4 CPU 2.40GHz stepping: 7 cpu MHz : 2392.077 cache size : 512 KB fdiv_bug: no hlt_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 2 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm bogomips: 4771.02 slave:~# free total used free sharedbuffers cached Mem: 31051042355364 749740 04401514104 -/+ buffers/cache: 8408202264284 Swap: 779144 428072 351072 MASTER master:~# cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 15 model : 2 model name : Intel(R) Pentium(R) 4 CPU 2.40GHz stepping: 7 cpu MHz : 2392.163 cache size : 512 KB fdiv_bug: no hlt_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 2 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm bogomips: 4771.02 master:~# free
Re: Replication suddenly stops on mysql 4.1.7 with Slave_IO_Running: No
Hello. I've looked through the bug database, and the only thing that I've found was an already-closed bug: http://bugs.mysql.com/bug.php?id=6148 Check that your server passes rpl_relayspace.test. Go to the mysql-test directory and execute: ./mysql-test-run t/rpl_relayspace.test Jan Kirchhoff [EMAIL PROTECTED] wrote: Hi, My problem still goes on... After having had the problem 2 more times within 1 day, I decided to re-do the replication (copy the whole database onto the slave with rsync and reset master and slave). That only lasted for little more than 1 day and I ended up with the same error: Could not parse relay log event entry. The possible reasons are: the master's binary log is corrupted (you can check this by running 'mysqlbinlog' on the binary log), the slave's relay log is corrupted (you can check this by running 'mysqlbinlog' on the relay log), a network problem, or a bug in the master's or slave's MySQL code. If you want to check the master's binary log or slave's relay log, you will be able to know their names by issuing 'SHOW SLAVE STATUS' on this slave. I can look at the binlog with mysqlbinlog on the master and the slave; no errors or problems. After a simple SLAVE START without having done any changes to the database, the slave thread startet again and caught up with the master. I've been using mysql's replication-feature since it first came up in 1999 or 2000 and dealt with lots of problems and workarounds, but this one is weird. Any ideas anybody? Jan Hello, I have a replication setup on to linux boxes (debian woody, kernel 2.4.21-xfs, mysql 4.1.7-standard official intel-compiler binary from mysql.com). master:~# mysqladmin status Uptime: 464848 Threads: 10 Questions: 296385136 Slow queries: 1752 Opens: 2629 Flush tables: 1 Open tables: 405 Queries per second avg: 637.596 slave:~# mysqladmin status Uptime: 463460 Threads: 2 Questions: 292885156 Slow queries: 6 Opens: 2510 Flush tables: 1 Open tables: 327 Queries per second avg: 631.953 both systems have identical hardware (P4 2.4ghz, 3GB RAM, SCSI-Hardware-RAID) connection is gigabit-ethernet. Everything used to work fine, but I wanted to get rid of InnoDB since I did only use that for very big table containing historical data and those tables were moved to another server. I ran out of discspace, innodb-datafiles can only grow but not shrink and i didn't need it anyway, so it had to go. I stopped the slave, changed all left over innodb-tables to myisam, added skip-innodb to my.cnf on the master and the slave, restarted the server, renewed the replication by doing it the classical way: flush tables with read log, copy the /var/lib/mysql on the slave (not much, just around 20GB), reset master, unlock tables. Then start the slave-mysqld, reset slave, slave start. Everything was fine and very fast for 4 days (from saturday till wednesday afternoon), then suddenly the slave stopped. this is where the weird stuff starts: show slave status tells me everything is fine, just Slave_IO_Running: No is wrong. After typing slave start, it says Slave_IO_Running: Yes, and Slave_SQL_Running: No. Very strange. Now i did a slave stop;slave start; and everything is fine again, the slave catches up and goes on. Today (thursday afternoon), the same thing happens again and can be solved again by slave stop;slave start;. Now it happened again around 10pm. Again, the stop-start-trick made it working again. I add the output of my mysql-shell Can anybody help me with that? This is a production system under heavy load and I can't play around with different mysql-versions and such... If I don't find a solution really quick, I'll have to do help myself with some shell-skript-daemon checking if replication is running and issuing stop slave;start slave-commands otherwise... not really the way it should be :( Thanks Jan SLAVE: slave:~# cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 15 model : 2 model name : Intel(R) Pentium(R) 4 CPU 2.40GHz stepping: 7 cpu MHz : 2392.077 cache size : 512 KB fdiv_bug: no hlt_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 2 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm bogomips: 4771.02 slave:~# free total used free sharedbuffers cached Mem: 31051042355364 749740 04401514104 -/+ buffers/cache: 8408202264284 Swap: 779144 428072 351072 MASTER master:~# cat
Replication suddenly stops on mysql 4.1.7 with Slave_IO_Running: No
Hello, I have a replication setup on to linux boxes (debian woody, kernel 2.4.21-xfs, mysql 4.1.7-standard official intel-compiler binary from mysql.com). master:~# mysqladmin status Uptime: 464848 Threads: 10 Questions: 296385136 Slow queries: 1752 Opens: 2629 Flush tables: 1 Open tables: 405 Queries per second avg: 637.596 slave:~# mysqladmin status Uptime: 463460 Threads: 2 Questions: 292885156 Slow queries: 6 Opens: 2510 Flush tables: 1 Open tables: 327 Queries per second avg: 631.953 both systems have identical hardware (P4 2.4ghz, 3GB RAM, SCSI-Hardware-RAID) connection is gigabit-ethernet. Everything used to work fine, but I wanted to get rid of InnoDB since I did only use that for very big table containing historical data and those tables were moved to another server. I ran out of discspace, innodb-datafiles can only grow but not shrink and i didn't need it anyway, so it had to go. I stopped the slave, changed all left over innodb-tables to myisam, added skip-innodb to my.cnf on the master and the slave, restarted the server, renewed the replication by doing it the classical way: flush tables with read log, copy the /var/lib/mysql on the slave (not much, just around 20GB), reset master, unlock tables. Then start the slave-mysqld, reset slave, slave start. Everything was fine and very fast for 4 days (from saturday till wednesday afternoon), then suddenly the slave stopped. this is where the weird stuff starts: show slave status tells me everything is fine, just Slave_IO_Running: No is wrong. After typing slave start, it says Slave_IO_Running: Yes, and Slave_SQL_Running: No. Very strange. Now i did a slave stop;slave start; and everything is fine again, the slave catches up and goes on. Today (thursday afternoon), the same thing happens again and can be solved again by slave stop;slave start;. Now it happened again around 10pm. Again, the stop-start-trick made it working again. I add the output of my mysql-shell Can anybody help me with that? This is a production system under heavy load and I can't play around with different mysql-versions and such... If I don't find a solution really quick, I'll have to do help myself with some shell-skript-daemon checking if replication is running and issuing stop slave;start slave-commands otherwise... not really the way it should be :( Thanks Jan SLAVE: slave:~# cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 15 model : 2 model name : Intel(R) Pentium(R) 4 CPU 2.40GHz stepping: 7 cpu MHz : 2392.077 cache size : 512 KB fdiv_bug: no hlt_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 2 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm bogomips: 4771.02 slave:~# free total used free sharedbuffers cached Mem: 31051042355364 749740 04401514104 -/+ buffers/cache: 8408202264284 Swap: 779144 428072 351072 MASTER master:~# cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 15 model : 2 model name : Intel(R) Pentium(R) 4 CPU 2.40GHz stepping: 7 cpu MHz : 2392.163 cache size : 512 KB fdiv_bug: no hlt_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 2 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm bogomips: 4771.02 master:~# free total used free sharedbuffers cached Mem: 31051043096016 9088 06482087780 -/+ buffers/cache:10075882097516 Swap: 779144 391732 387412 Slave shell: wpdb2:~# mysql Welcome to the MySQL monitor. Commands end with ; or \g. Your MySQL connection id is 23083 to server version: 4.1.7-standard Type 'help;' or '\h' for help. Type '\c' to clear the buffer. wpdb2 mysql show slave status\G *** 1. row *** Slave_IO_State: Master_Host: 192.168.10.26 Master_User: repl Master_Port: 3306 Connect_Retry: 10 Master_Log_File: mysql-bin.000210 Read_Master_Log_Pos: 146168522 Relay_Log_File: wpdb2-relay-bin.000210 Relay_Log_Pos: 146168608 Relay_Master_Log_File: mysql-bin.000210 Slave_IO_Running: No Slave_SQL_Running: Yes Replicate_Do_DB: Replicate_Ignore_DB: Replicate_Do_Table: Replicate_Ignore_Table: Replicate_Wild_Do_Table: Replicate_Wild_Ignore_Table: