[Gllug] Mysql problem

[Alain Williams]

Hi,

Summary: mysqld segmentation fault -- DNS related ?

I have mysqld running on a RedHat 7.3 box, clients local to the box and from M$ Windows (ODBC).
It has been stable for quite some time. I am up to date with all the latest patches.

Today when a windows client attempts to connect, the server segfaults.

Below is the output of strace on one of the mysqld (servers) at the time that the client tries to connect.
The client's IP address was not in DNS (it never has been), I added it just in case, but that made no difference.
I added the IP address to /etc/hosts and the client can connect, etc.
In /etc/nsswitch.conf I have: ``hosts: files dns''


Anyone any idea what is happening ?

No, I don't want to add all the PCs into /etc/hosts.

9365  open("/etc/ld.so.cache", O_RDONLY) = 8
9365  fstat64(8, {st_mode=S_IFREG|0644, st_size=77335, ...}) = 0
9365  old_mmap(NULL, 77335, PROT_READ, MAP_PRIVATE, 8, 0) = 0x40907000
9365  close(8)                          = 0
9365  open("/lib/libnss_dns.so.2", O_RDONLY) = 8
9365  read(8, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0\320\r\0"..., 1024) = 1024
9365  fstat64(8, {st_mode=S_IFREG|0755, st_size=16051, ...}) = 0
9365  old_mmap(NULL, 15912, PROT_READ|PROT_EXEC, MAP_PRIVATE, 8, 0) = 0x4001e000
9365  mprotect(0x40021000, 3624, PROT_NONE) = 0
9365  old_mmap(0x40021000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 8, 0x2000) = 0x40021000
9365  close(8)                          = 0
9365  open("/lib/libresolv.so.2", O_RDONLY) = 8
9365  read(8, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0\20\'\0"..., 1024) = 1024
9365  fstat64(8, {st_mode=S_IFREG|0755, st_size=68925, ...}) = 0
9365  old_mmap(NULL, 69408, PROT_READ|PROT_EXEC, MAP_PRIVATE, 8, 0) = 0x40924000
9365  mprotect(0x40932000, 12064, PROT_NONE) = 0
9365  old_mmap(0x40932000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 8, 0xe000) = 0x40932000
9365  old_mmap(0x40933000, 7968, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x40933000
9365  close(8)                          = 0
9365  munmap(0x40907000, 77335)         = 0
9365  --- SIGSEGV (Segmentation fault) ---
6889  <... poll resumed> [{fd=5, events=POLLIN}], 1, 2000) = -1 EINTR (Interrupted system call)
6889  --- SIGRT_1 (Real-time signal 1) ---
6889  sigreturn()                       = ? (mask now ~[TRAP KILL STOP])
6889  getppid()                         = 6886
6889  wait4(-1, [WIFSIGNALED(s) && WTERMSIG(s) == SIGSEGV], WNOHANG|__WCLONE, NULL) = 9365
6889  modify_ldt(0x1, 0x81e7cdc, 0x10)  = 0
6889  munmap(0x408f6000, 69632)         = 0
6889  kill(6890, SIGSEGV)               = 0
6889  kill(6886, SIGSEGV)               = 0
6889  _exit(0)                          = ?

The 2 other process killed are the other mysqlds that are running - one of which does not die.

-- 
Alain Williams

#include <std_disclaimer.h>

-- 
Gllug mailing list  -  Gllug at linux.co.uk
http://list.ftech.net/mailman/listinfo/gllug