[Bug 675185] Re: [Hardy SRU] dash bug causes mysqld_safe to spin at 100% CPU

Sun Nov 14 14:04:14 GMT 2010

** Patch added: "fix.diff"
   https://bugs.launchpad.net/bugs/675185/+attachment/1733135/+files/fix.diff

** Description changed:

  Symptoms are: After mysql server upgrades + restarts the old mysqld_safe keeps running and consumes 100% cpu, strace shows:
- 14:18:18.789847 wait4(4294967295, 0x7fffb3b3d59c, 0, NULL) = -1 ECHILD (No child processes)
- 14:18:18.789920 wait4(4294967295, 0x7fffb3b3d59c, 0, NULL) = -1 ECHILD (No child processes)
- 14:18:18.789992 wait4(4294967295, 0x7fffb3b3d59c, 0, NULL) = -1 ECHILD (No child processes)
+ wait4(4294967295, 0x7fffb3b3d59c, 0, NULL) = -1 ECHILD (No child processes)
+ wait4(4294967295, 0x7fffb3b3d59c, 0, NULL) = -1 ECHILD (No child processes)
+ wait4(4294967295, 0x7fffb3b3d59c, 0, NULL) = -1 ECHILD (No child processes)
  For more info see the dupes, particularily bug 105457. Does not happen always, for reliable test case see below.

  This is a bug (race condition) in dash, fixed by
  http://git.kernel.org/?p=utils/dash/dash.git;a=commit;h=3800d4934391b144fd261a7957aea72ced7d47ea
  .

  TEST CASE: Script in the commit message (link above). Allows to
  reproduce this always, even on Karmic still.

  SRU JUSTIFICATION: Obviously safe patch.

  PATCH: Attached, run mysqld_safe in bash instead of dash.

  REGRESSION POTENTIAL: Minimal. 1) Bash is a superset of dash. 2) Before
  dash became the default sh, bash was used, we just revert that change.

  NOTES: mysqld_safe will be replaced by an upstart job. But Hardy has
  support for 3 more years, justifiying a fix.

** Description changed:

  Symptoms are: After mysql server upgrades + restarts the old mysqld_safe keeps running and consumes 100% cpu, strace shows:
- wait4(4294967295, 0x7fffb3b3d59c, 0, NULL) = -1 ECHILD (No child processes)
- wait4(4294967295, 0x7fffb3b3d59c, 0, NULL) = -1 ECHILD (No child processes)
- wait4(4294967295, 0x7fffb3b3d59c, 0, NULL) = -1 ECHILD (No child processes)
+ wait4(4294967295, 0x7fffb3b3d59c, 0, NULL) = -1 ECHILD (No child ...
+ wait4(4294967295, 0x7fffb3b3d59c, 0, NULL) = -1 ECHILD (No child ...
+ wait4(4294967295, 0x7fffb3b3d59c, 0, NULL) = -1 ECHILD (No child ...
  For more info see the dupes, particularily bug 105457. Does not happen always, for reliable test case see below.