| Bug #31123 | InnoDB: Warning: purge reached the head of the history list | ||
|---|---|---|---|
| Submitted: | 21 Sep 2007 2:50 | Modified: | 13 Feb 2011 0:56 |
| Reporter: | Bingxi Wu | Email Updates: | |
| Status: | Closed | Impact on me: | |
| Category: | MySQL Server: InnoDB storage engine | Severity: | S2 (Serious) |
| Version: | 5.0.45, 5.1 | OS: | Any |
| Assigned to: | Sunny Bains | CPU Architecture: | Any |
[21 Sep 2007 4:20]
Valeriy Kravchuk
Thank you for a problem report. Please, send your my.cnf. Is there any indication of hardware-related (RAM) problems?
[21 Sep 2007 5:18]
Bingxi Wu
my.cnf for innodb variables innodb_file_per_table = 1 innodb_support_xa = 0 innodb_status_file = 1 innodb_buffer_pool_size = 2048M innodb_additional_mem_pool_size = 8M innodb_log_file_size = 256M innodb_log_buffer_size = 16M innodb_lock_wait_timeout = 100 innodb_flush_log_at_trx_commit = 0 innodb_flush_method = 'O_DIRECT' set-variable="transaction-isolation=READ-COMMITTED" innodb_file_io_threads = 4 innodb_thread_concurrency = 16 innodb_log_files_in_group = 3 innodb_max_dirty_pages_pct = 90 Hardware: Dell 2095 5310 Add 6G mem and 2 SAS 300G disk So, 8G mem and 3 SAS 300G make RIAD 5 4 core cpu 1.6 Cache 8M dmesg: CPU0: Physical Processor ID: 0 CPU0: Processor Core ID: 0 CPU0: Initial APIC ID: 0 Using IO APIC NMI watchdog CPU: L1 I cache: 32K, L1 D cache: 32K CPU: L2 cache: 4096K CPU0: Physical Processor ID: 0 CPU0: Processor Core ID: 0 CPU0: Initial APIC ID: 0 CPU0: Intel(R) Xeon(R) CPU E5310 @ 1.60GHz stepping 07 per-CPU timeslice cutoff: 4094.43 usecs. task migration cache decay timeout: 4 msecs. Booting processor 1/1 rip 6000 rsp 10037eb5f58 Initializing CPU#1 Calibrating delay using timer specific routine.. 3192.09 BogoMIPS (lpj=1596047) CPU: L1 I cache: 32K, L1 D cache: 32K CPU: L2 cache: 4096K CPU1: Physical Processor ID: 0 CPU1: Processor Core ID: 1 CPU1: Initial APIC ID: 1 Intel(R) Xeon(R) CPU E5310 @ 1.60GHz stepping 07 Booting processor 2/2 rip 6000 rsp 1000696df58 Initializing CPU#2 Calibrating delay using timer specific routine.. 3192.11 BogoMIPS (lpj=1596058) CPU: L1 I cache: 32K, L1 D cache: 32K CPU: L2 cache: 4096K CPU2: Physical Processor ID: 0 CPU2: Processor Core ID: 2 CPU2: Initial APIC ID: 2 Intel(R) Xeon(R) CPU E5310 @ 1.60GHz stepping 07 Booting processor 3/3 rip 6000 rsp 10228523f58 Initializing CPU#3 Calibrating delay using timer specific routine.. 3192.11 BogoMIPS (lpj=1596058) CPU: L1 I cache: 32K, L1 D cache: 32K CPU: L2 cache: 4096K CPU3: Physical Processor ID: 0 CPU3: Processor Core ID: 3 CPU3: Initial APIC ID: 3 Intel(R) Xeon(R) CPU E5310 @ 1.60GHz stepping 07 Total of 4 processors activated (12770.95 BogoMIPS).
[21 Sep 2007 14:45]
Heikki Tuuri
I have seen several reports of this over years. This is a mild form of corruption, but its cause is a mystery.
[21 Sep 2007 15:46]
Bingxi Wu
Thank you Heikki Tuuri . you mean my innodb space was corruption. But i don't know How to repair it . Can you give me some pint make it well. Now ,i only want to know how to make the Innodb Good. Don't out print the message. How can I do?
[24 Sep 2007 16:49]
Heikki Tuuri
I think the only way to remove the message is to recreate the whole InnoDB tablespace :(.
[25 Sep 2007 1:46]
Bingxi Wu
Thank you Heikki Tuuri. recent i can't to recreate innodb tablespace. The database Very big .and can't shutdown mysql.
[28 Nov 2007 21:18]
James Day
Looks like a duplicate of bug #23552.
[14 Dec 2007 16:00]
Clifton Kilby III
I would just like to chime in that: 071214 10:41:33 InnoDB: Warning: purge reached the head of the history list, InnoDB: but its length is still reported as 3568147! Make a detailed bug InnoDB: report, and post it to bugs.mysql.com We run MySQL 5.0.37-log Source distribution on Dell_RHEL4a Linux 2.6.9-5.ELsmp MemTotal: 4151368 kB Intel(R) Xeon(TM) CPU 2.80GHz X 4 This is a production environment, and it is necessary that we use a transactionable data storage system. Its been running fine for 6 months now, with god only knows how much traffic, so I am not too worried, but I see a lot of "corrupt" and "invalid" in articles regarding this. The only reason I found this was during a fs cleanup to reclaim diskspace I noticed that mysql was complaining, a lot. Approximately 47M worth.
[24 Jun 2008 16:24]
Heikki Tuuri
This corruption bug might happen if one uses wrong ib_logfiles to recover from a mysqld crash. Then the structure of the history list may be fail to be redone in crash recovery. Unfortunately, the bug reports do not support this hypothesis yet. We need the .err log from a long time to check this.
[26 Jun 2008 5:24]
Matthew Clark
I started having the same mysqld.log warnings (InnoDB: Warning: purge reached the head of the history list). I have upgraded on Fedora 6 from 5.0.22 to 5.0.51a and the warnings persist. I have done a complete mysqldump and restore. Any suggestions for how to clean up this problem? I can send more information if needed. Thanks in advance, Matthew Clark
[1 Dec 2008 19:28]
Mikhail Izioumtchenko
could we have mysqld error log from the latest occurrence, if this bug is till a problem?
[2 Jan 2009 0:00]
Bugs System
No feedback was provided for this bug for over a month, so it is being suspended automatically. If you are able to provide the information that was originally requested, please do so and change the status of the bug back to "Open".
[8 Jan 2009 14:11]
Tadas Z
# grep 'InnoDB: Warning: purge reached the head of the history list,' mysqld.log |wc -l 2070 # mysql --version mysql Ver 14.12 Distrib 5.0.45, for redhat-linux-gnu (x86_64) using readline 5.0 (centos5 x64) i've seen this occuring on several hosts here around, so a resolution is still needed.
[11 Jan 2009 12:57]
Sveta Smirnova
Tadas, thank you for the feedback. But version 5.0.45 is old. Please upgrade to current version 5.0.67, try with it and if problem still occurs upload full error log independent from its size. You can upload it on our FTP server as described in "Files" tab of the bug report.
[12 Feb 2009 0:00]
Bugs System
No feedback was provided for this bug for over a month, so it is being suspended automatically. If you are able to provide the information that was originally requested, please do so and change the status of the bug back to "Open".
[3 Apr 2009 13:40]
Darren Birkett
[root@xxxxxx ~]# grep 'InnoDB: Warning: purge reached the head of the history list,' /var/log/mysqld.log|wc -l 66 [root@xxxxxx ~]# mysql --version mysql Ver 14.12 Distrib 5.0.67, for redhat-linux-gnu (x86_64) using readline 5.1
[2 Feb 2010 0:06]
MySQL Verification Team
See http://bugs.mysql.com/bug.php?id=50772.
[6 Apr 2010 21:10]
Calvin Sun
Bug#50772 is marked as a dup of this one.
[7 Jan 2011 18:07]
Calvin Sun
bug#54965 is marked as a dup of this one. And change the status to Verified (since 54965 is verified) even it is hard to reproduce.
[13 Feb 2011 0:56]
Bingxi Wu
innodb space was corruption recreate the whole InnoDB tablespace
Description: there is some error with Mysql Verson: mysql-5.0.45 tail -f error.log InnoDB: but its length is still reported as 117019! Make a detailed bug InnoDB: report, and post it to bugs.mysql.com 070921 10:28:40 InnoDB: Warning: purge reached the head of the history list, InnoDB: but its length is still reported as 117021! Make a detailed bug InnoDB: report, and post it to bugs.mysql.com 070921 10:28:50 InnoDB: Warning: purge reached the head of the history list, InnoDB: but its length is still reported as 117026! Make a detailed bug InnoDB: report, and post it to bugs.mysql.com 070921 10:29:00 InnoDB: Warning: purge reached the head of the history list, InnoDB: but its length is still reported as 117028! Make a detailed bug InnoDB: report, and post it to bugs.mysql.com 070921 10:29:10 InnoDB: Warning: purge reached the head of the history list, InnoDB: but its length is still reported as 117031! Make a detailed bug InnoDB: report, and post it to bugs.mysql.com 070921 10:29:20 InnoDB: Warning: purge reached the head of the history list, InnoDB: but its length is still reported as 117036! Make a detailed bug InnoDB: report, and post it to bugs.mysql.com 070921 10:29:30 InnoDB: Warning: purge reached the head of the history list, InnoDB: but its length is still reported as 117048! Make a detailed bug InnoDB: report, and post it to bugs.mysql.com 070921 10:29:42 InnoDB: Warning: purge reached the head of the history list, InnoDB: but its length is still reported as 117051! Make a detailed bug InnoDB: report, and post it to bugs.mysql.com 070921 10:29:52 InnoDB: Warning: purge reached the head of the history list, InnoDB: but its length is still reported as 117056! Make a detailed bug InnoDB: report, and post it to bugs.mysql.com 070921 10:30:02 InnoDB: Warning: purge reached the head of the history list, InnoDB: but its length is still reported as 117061! Make a detailed bug InnoDB: report, and post it to bugs.mysql.com 070921 10:30:12 InnoDB: Warning: purge reached the head of the history list, InnoDB: but its length is still reported as 117066! Make a detailed bug InnoDB: report, and post it to bugs.mysql.com uname -a Linux db198 2.6.9-55.ELsmp #1 SMP Wed May 2 14:04:42 EDT 2007 x86_64 x86_64 x86_64 GNU/Linux free total used free shared buffers cached Mem: 8166012 8146480 19532 0 581952 3164824 -/+ buffers/cache: 4399704 3766308 Swap: 8193140 208 8192932 getconf GNU_LIBPTHREAD_VERSION NPTL 2.3.4 What Should I do? How to repeat: Use Mysql source build As slave