[dev.icinga.com #2666] various memory leaks in downtime eventhandling on SIGHUP

[icinga-migration]

This issue has been migrated from Redmine: https://dev.icinga.com/issues/2666

Created by mfriedrich on 2012-06-06 10:46:08 +00:00

Assignee: mfriedrich
Status: Resolved (closed on 2012-06-18 11:52:25 +00:00)
Target Version: 1.7.1
Last Update: 2012-06-18 11:52:25 +00:00 (in Redmine)

---

needs proper valgrind'ing, with embedded perl enabled.

Attachments

• nagios-3.4.1-leaks1-2.patch mfriedrich - 2012-06-06 10:46:08 +00:00

Changesets

2012-06-15 12:37:34 +00:00 by mfriedrich 7383da1

core: fix various memory leaks in downtime eventhandling on SIGHUP (Carlos Velasco) #2666

under various circumstances, the event_data was not cleared correctly
when reloading the core, therefore causing memory leaks.

refs #2666

[icinga-migration]

Updated by mfriedrich on 2012-06-15 12:30:26 +00:00

• Subject changed from various possible memory leaks to various memory leaks in downtime eventhandling on SIGHUP

• Target Version changed from 1.8 to 1.7.1

  core: Plug leaks in downtime eventhandling on SIGHUP
  When we receive SIGHUP and downtime events are scheduled for later,
  we would previously leak the event data for those downtimes.

  This patch supposedly fixes it (and it certainly looks as if it
  does), but more testing on that matter would be welcome.

  To test it:

  • run Nagios through valgrind --leak-check=full
  • schedule downtime to happen in the future
  • Restart Nagios in a multitude of different ways
    • by sending RESTART_PROCESS to command-pipe
    • by sending SIGHUP via kill(1)
    • by restarting outright (stop + start)

  Patched-by: Carlos Velasco
  Signed-off-by: Andreas Ericsson

[icinga-migration]

Updated by mfriedrich on 2012-06-18 11:52:25 +00:00

• Status changed from Assigned to Resolved
• Done % changed from 0 to 100