Bug #112148 Description Failed to upgrade mysql from 8.0.20 to 8.0.34
Submitted: 23 Aug 2023 3:34 Modified: 23 Aug 2023 12:53
Reporter: qifx qi Email Updates:
Status: Can't repeat Impact on me:
None 
Category:MySQL Server Severity:S3 (Non-critical)
Version:8.0.34 OS:Red Hat (Red Hat Enterprise Linux Server 7.7(Maipo))
Assigned to: CPU Architecture:x86 (Hygon C86 7375)
Tags: 8.0.34, failed, Hygon, redhat, upgrade

[23 Aug 2023 3:34] qifx qi 
Description:
The following error occurred when I upgraded mysql from 8.0.20 to 8.0.34:
2023-08-18T01:56:46.201651Z 0 [System]  [MY-013951] [Server]  BuildID[sha1]=f7f7356a527b9ed410773c69291cedce564eede9
2023-08-18101:56:46.201652Z 0 [System]  [MY-013951] [Server]  Thread pointer: 0x0
2023-08-18T01:56:46.201653Z 0 [System]  [MY-013951] [Server]  Attempting backtrace. You can use the following information to find out
2023-08-18T01:56:46.201654Z 0 [System]  [MY-013951] [Server]  where mysqld died.If you see no messages after this, something went
2023-08-18T01:56:46.201655Z 0 [System]  [MY-013951] [Server]  terribly wrong...
2023-08-18T01:56:46.201656Z 0 [System]  [MY-013951] [Server]  stack bottom = 0 thread_stack 0x100000
2023-08-18T01:56:46.201657Z 0 [System]  [MY-013951] [Server]  /opt/mysql/bin/mysqld(my_print_stacktrace(unsigned char const*,unsigned long)+0x2e) [0x20a03be]
2023-08-18T01:56:46.201658Z 0 [System]  [MY-013951] [Server]  /opt/mysql/bin/mysqld(print_fatal_signal(int)+0x3c3) [0x112c933]
2023-08-18T01:56:46.201659Z 0 [System]	[MY-013951]	[Serverl	/datacanva8/dep/mysq1/bin/mysqld(handle_fatal_signal+0x65)[0x112c9a5]	
2023-08-18T01:56:46.201660Z 0 [System]	[MY-0139511	[Server]  /1ib64/1ibpthread.so.0(+0xf630)[0x7fe862324630]	
2023-08-18701:56146,201661Z 0 [System]	[MY-013951]	[Server]  /opt/mysq1/bin/mysqld(Delegate::Delegate(unsigned int)+0xad) [0x1d316fd]	
2023-08-18701:56:46.201662Z 0 [System]	[MY-013951]	[Server]  /opt/mysq1/bin/mysqld(delegates_init()+0x20)[0xld31c10]
2023-08-18T01:56:46.201663Z 0 [System]	[MY-013951]	[Server]  /opt/mysg1/bin/mysq1d()[0xeb87d5]	
2023-08-18701:56:46.201664Z 0 [System]	[MY-0139511	[Server]  /opt/mysq1/bin/mysqld(mysq1d_main(int，char**)+0x2d42)[0xec23b2]
2023-08-18T01:56:46.201665Z 0 [System]	[MY-013951] [server]  /1ib64/1ibc.so.6(__11bc_start_main+0xf5) [0x7fe8603d4545]
2023-08-18701:56:46.201666Z 0 [System]  [MY-013951]	[Server]  /opt/mysq1/bin/mysq1d() [0xea28f5]
2023-08-18T01:56:46.201667Z 0 [System]  [MY-013951] [Server]  The manual page at http://dev.mysql.com/doc/myaq1/en/crashing.html contains	
2023-08-18T01:56:46.201668Z 0 [System]  [MY-0139511 [Server]  information that should help you find out what is causing the crash.	
2023-08-18T01:56:46.968026Z 0 [Warning]  [MY-011070] [Server] 'Disabling symbolic links using --skip-symbolic-links (or equivalent) is the default. Consider not using this option as it’ is deprecated and will be removed in a future release.		
2023-08-18T01:56:46.968064Z 0 [Warning]  [MY-011068] [Server] The syntax 'expire-logs-days is deprecated and will be removed in a future release.  Please use binlog_expire_logs_seconds instead.
2023-08-18T01:56146.968292Z 0 [System]  [MY-010116] [server]  /opt/mysql/bin/mysqd (mysqld 8.0.34) starting as process 66559	
2023-08-18T01:56:46.974215Z 0 [System]	[MY-013951] [Serverl  A bacitrace was processed and added to the main error-log in the appropriate format.	
2023-08-18T01:56:462 UTC - mysqld got signal 11 ;	
Most likely, you have hit a bug, but this error can also be caused by malfunctioning hardware.	
BuildID[shal]=f7f7356a527b9ed410773069291cedce564eede9
Thread pointer: 0x0	
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 = 0 thread_stack 0x100000
/opt/mysql/bin/mysqld(my_print_stacktrace(unsigned char const', unsigned long)+0x2e) [0x20a03be]
/opt/mysql/bin/myaqld(print_fatal_signal(int)+0x3c3) [0x112c933]	
/opt/mysql/bin/mysqld(handle_fatal_signal+0x65 (0x112c9a5]	
/1ib64/libpthread.so.0(+0xf630) [0x7f12f19a4630]
/opt/mysql/bin/mysqld(Delegate::Delegate(unsigned int)+0xad)[0x1d316fd]	
/opt/mysql/bin/mysqld(delegates_init()+0x20) [0x1d31c10]	
/opt/mysql/bin/mysqld() [0xeb87d5]	
/opt/mysql/bin/mysqld(mysqd_main(int,cha'')0x2d42) [0xec23b2]
/1ib64/11bc.s0.6(__libc_start_main+0xf5)(0x7f12efa54545]
/opt/mysql/bin/mysqld() [0xea2855]
The manual page at http://dev.mysql.com/doc/mysql/en/crashing.html contains

This problem occurs when I upgrade from 8.0.20 to 8.0.34 and when I install 8.0.34. The same error occurs when I install in glibc mode as when I install in rpm. My system glibc version is gblic-2.17

How to repeat:
It can be repeated when upgrading from 8.0.20 to 8.0.34 and when installing new 8.0.34, regardless of whether the glibc installation is in glibc mode or rpm mode, glibc version is glibC-2.17
[23 Aug 2023 12:53] MySQL Verification Team 
Hi Mr. qi,

Thank you so much for your bug report.

However, we were not able to repeat it. Our MySQL upgrade on Red Hat 7 had no problems what so ever.

Hence, there are several possibilities. 

First one is that your bug is a duplicate of this verified bug:

https://bugs.mysql.com/bug.php?id=110810

That bug is private, so you can't read it. Hence, we have to ask you some questions.

Our first question is the following one:

Have you tried downloading a proper RPM for MySQL  and then trying running the upgrade. Our RPM packages are on dev.mysql.com. There are packages for many versions of Red Hat Enterprise Linux. Please, try downloading and installing the one for the RH version 7 !!!!!!

Next ......

Let us know, what kind of CPU is Hygon C86 7375 ???? Is it Intel compatible ???

Would be possible for you to post the output of the following shell command?

getconf LEVEL1_DCACHE_LINESIZE

Next, can you run the following commands:

# grep cache_alignment /proc/cpuinfo

# cat /sys/devices/system/cpu/cpu0/cache/*/coherency_line_size

Maybe there's a bug in glibc, which  the outputs from these commands would reveal.

Next, please write the following program and name it test.c:

#include <unistd.h>
#include <stdio.h>

int main(int argc, char *argv[])
{
    int cacheLineSize = sysconf(_SC_LEVEL1_DCACHE_LINESIZE);
    printf("LEVEL1_DCACHE_LINESIZE: %d\n", cacheLineSize);

    return 0;
}

Next, build the test program, as the user root,  like this:

# gcc -o  test test.c

Then run it as user root, like this:

# ./test

Send us also the output from that program.

Next, if your CPU is from AMD, let us know whether does your CPU happen to be an older AMD CPU where cpuid only returns cache details in the 0x80000005 leaf but not in the 0x8000001d leaf?

In order to find out that, run program `cpuid` with no options, then with all other options, one by one, including `cpuid -1` and `cpuid -r`.

The above feedback would be fine as a start.

However, do not forget to use a proper RPM for RH 7 from our GPL download pages.