I just ran a verification on this package.

root@juju-niedbalski-sec-machine-27:/home/ubuntu# for file in $(strace -e open 
-i corosync -v 2>&1 | grep -E '.*shm.*' |grep -Po '".*?"'| sed -e s/\"//g); do 
du -sh $file; done
12K     /dev/shm/qb-corosync-blackbox-header
8.1M    /dev/shm/qb-corosync-blackbox-data

After enabling proposed

root@juju-niedbalski-sec-machine-27:/home/ubuntu# dpkg -l | grep corosync
ii  corosync                         2.3.3-1ubuntu2                   amd64     
   Standards-based cluster framework (daemon and modules)
ii  libcorosync-common4              2.3.3-1ubuntu2                   amd64     
   Standards-based cluster framework, common library

root@juju-niedbalski-sec-machine-27:/home/ubuntu# for file in $(strace -e open 
-i corosync -v 2>&1 | grep -E '.*shm.*' |grep -Po '".*?"'| sed -e s/\"//g); do 
du -sh $file; done
du: cannot access ‘/dev/shm/qb-corosync-blackbox-header’: No such file or 
directory
du: cannot access ‘/dev/shm/qb-corosync-blackbox-data’: No such file or 
directory

So, seems to be fixed.

** Tags removed: verification-needed
** Tags added: verification-done

-- 
You received this bug notification because you are a member of Ubuntu
High Availability Team, which is subscribed to corosync in Ubuntu.
https://bugs.launchpad.net/bugs/1530837

Title:
  Logsys file leaks in /dev/shm after sigabrt, sigsegv and when running
  corosync -v

Status in corosync package in Ubuntu:
  Fix Released
Status in corosync source package in Trusty:
  Fix Committed

Bug description:
  [Impact]

   * corosync has a memory leak problem with multiple calls to corosync -v
   * corosync has a memory leak problem by not properly handling signals

  [Test Case]

   * run "corosync -v" multiple times
   * some cloud tools do that

  [Regression Potential] 
   
   * minor code changes on not-core code
   * based on upstream changes
   * based on a redhat fix

  [Other Info]

  # Original BUG Description

  It was brought to my attention that Ubuntu also suffers from:

  https://bugzilla.redhat.com/show_bug.cgi?id=1117911

  And corosync should include the following fixes:

  ----

  commit dfaca4b10a005681230a81e229384b6cd239b4f6
  Author: Jan Friesse <[email protected]>
  Date: Wed Jul 9 15:52:14 2014 +0200

  Fix compiler warning introduced by previous patch

  QB loop signal handler prototype differs from signal(2) prototype.
  Solution is to create wrapper functions.

  Signed-off-by: Jan Friesse <[email protected]>

  commit 384760cb670836dc37e243f594612c6e68f44351
  Author: zouyu <[email protected]>
  Date: Thu Jul 3 10:56:02 2014 +0800

  Handle SIGSEGV and SIGABRT signals

  SIGSEGV and SIGABRT signals are now correctly handled (blackbox is
  dumped and logsys is finalized).

  Signed-off-by: zouyu <[email protected]>
  Reviewed-by: Jan Friesse <[email protected]>

  commit cc80c8567d6eec1d136f9e85d2f8dfb957337eef
  Author: zouyu <[email protected]>
  Date: Wed Jul 2 10:00:53 2014 +0800

  fix memory leak produced by 'corosync -v'

  Signed-off-by: zouyu <[email protected]>
  Reviewed-by: Jan Friesse <[email protected]>

  ----

  Description from Red Hat bug:

  """
  Description of problem:
  When corosync receives sigabrt or sigsegv it doesn't delete libqb blackbox 
file (/dev/shm one). Same happens when corosync is executed with -v parameter 
(this shows only version, so it shouldn't cause leak in /dev/shm).

  Version-Release number of selected component (if applicable):
  7.0

  How reproducible:
  100%

  Steps to Reproduce 1:
  1. Start corosync
  2. Send sigabrt to corosync

  Steps to Reproduce 1:
  1. Execute corosync -v

  Actual results:
  File like qb-corosync-*-blackbox-data|header exists results in leak of 
/dev/shm space.

  Expected results:
  No leak

  Additional info:
  """

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/corosync/+bug/1530837/+subscriptions

_______________________________________________
Mailing list: https://launchpad.net/~ubuntu-ha
Post to     : [email protected]
Unsubscribe : https://launchpad.net/~ubuntu-ha
More help   : https://help.launchpad.net/ListHelp

Reply via email to