[AMBARI-5784] Nagios read timeout on 1000 node cluster

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 1.6.1
• Component/s: ambari-server
• Assignee: Nate Cole
• Reporter: Nate Cole
• Created: 15/May/14 22:57
• Updated: 16/May/14 00:41
• Resolved: 16/May/14 00:41

Description

The following curl call takes forever,

curl -u admin:admin 'http://146.148.78.70:8080/api/v1/clusters/c1/services?fields=alerts/summary,ServiceInfo/state,ServiceInfo/maintenance_state&minimal_response=true&_=1398386567299'

real	5m55.246s
user	0m0.010s
sys	0m0.006s

Ambari server logs has the following exception:

00:50:24,980 ERROR [qtp1819107218-73] NagiosPropertyProvider:199 - Could not load metrics - Executor exception (org.apache.ambari.server.controller.spi.SystemException: Error reading HTTP response for cluster c1, nagios=http://perf-1000-b-4.c.pramod-thangali.internal/ambarinagios/nagios/nagios_alerts.php?q1=alerts&alert_type=all (Read timed out))

Read timeout should not be blocking call, although it cannot return anything.

Issue Links

• links to: ReviewBoard