[ZBX-16224] Possible semaphore leak under Solaris 10, Agent 4.2.1 Created: 2019 Jun 06  Updated: 2019 Jun 07  Resolved: 2019 Jun 07

Status: Closed
Project: ZABBIX BUGS AND ISSUES
Component/s: Agent (G)
Affects Version/s: 4.2.1
Fix Version/s: None

Type: Incident report Priority: Major
Reporter: David Angelovich Assignee: Zabbix Support Team
Resolution: Duplicate Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Solaris 10 8/11 X86


Attachments: File semaphore.log.gz     File zabbix_agentd.log.gz    
Issue Links:
Duplicate
duplicates ZBX-16126 Wrong number of selected triggers and... Closed

 Description   

Steps to reproduce:

  1. Build Zabbix Agent for Solaris 10
  2. Run agent unconfigured/configured - doesn't seem to make a difference whether its actually doing any real work or not.
  3. Run ipcs -s | grep zabbix periodically to watch the semaphore count grow until it maxes out
  4. Stop the agent
  5. Run ipcs -s | grep zabbix to check the count and notice only 1 was removed after exiting
  6. Run ipcrm to clean up semaphores

Result:
See attached zabbix agent log file, alongside a log of output from a loop that's counting the number of Zabbix semaphores:

for i in `seq 1 1000000`; do date >> /oanda/var/tmp/semaphore.log; ipcs -s | grep zabbix | wc -l >> /oanda/var/tmp/semaphore.log; sleep 1; done

 

Expected:
Semaphores were stable under 4.0.5, so I'm unclear when this changed exactly.



 Comments   
Comment by richlv [ 2019 Jun 07 ]

Likely a duplicate of ZBX-16216.

Comment by Arturs Lontons [ 2019 Jun 07 ]

Thank you for reporting the issue.

The ticket is a duplicate of ZBX-16126 and ZBX-16128. We're currently working on a fix for the issue. For now, I can recommend following ZBX-16128 for further comments from the Zabbix team regarding the upcoming fixes.

Generated at Fri Apr 26 07:30:11 EEST 2024 using Jira 9.12.4#9120004-sha1:625303b708afdb767e17cb2838290c41888e9ff0.