Bug 535822 (RHQ-2479)
Summary: | transaction timeout occurs when updating metric template and applying it to existing Resources, when there are a large number of Resources spread across many Agents | ||
---|---|---|---|
Product: | [Other] RHQ Project | Reporter: | Ian Springer <ian.springer> |
Component: | Monitoring | Assignee: | Jay Shaughnessy <jshaughn> |
Status: | CLOSED CURRENTRELEASE | QA Contact: | Heiko W. Rupp <hrupp> |
Severity: | medium | Docs Contact: | |
Priority: | high | ||
Version: | 1.3 | CC: | ccrouch, cwelton |
Target Milestone: | --- | ||
Target Release: | --- | ||
Hardware: | All | ||
OS: | All | ||
URL: | http://jira.rhq-project.org/browse/RHQ-2479 | ||
Whiteboard: | |||
Fixed In Version: | 2.4 | Doc Type: | Bug Fix |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2010-08-12 16:47:38 UTC | Type: | --- |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: | |||
Bug Depends On: | |||
Bug Blocks: | 557793 |
Description
Ian Springer
2009-10-29 15:01:00 UTC
NOTE: This issue was reported via support case 344497. The resource type that the customer was using was the Embedded Tomcat Connector type, of which they have around 5500 in inventory. the customer set the metric template up, hit OK and the UI spun for a while. They went to do something else, came back and saw the following error in the UI: java.lang.IllegalStateException: [com.arjuna.ats.internal.jta.transaction.arjunacore.inactive] [com.arjuna.ats.internal.jta.transaction.arjunacore.inactive] The transaction is not active! The page requested cannot be displayed due to some error. You can view the stack trace, return to the previous page, Dashboard, or Browse Resources page. (the full stack trace can be found in this JIRA's Description) This looks like a scalability issue due to the fact that we're utilizing attached objects for the schedule update logic. This is going to put severe load on the hibernate cache for large updates (lots of schedule definitions for large a large number of resources). As the cache gets large the performance goes down and timeouts can take place. The solution is most likely to eliminate the use of the attached objects and not put load on the cache. This would be done by performing direct db updates via query. Some care will be needed to also make sure the agent schedule updates are also performed efficiently. This bug was previously known as http://jira.rhq-project.org/browse/RHQ-2479 This bug is related to RHQ-390 Was there a code change made in for this? qa -> spinder (for perf env) NEEDINFO -> jshaughn for commit info. Yes, Jay checked in the fix for this in git rev 6acd49ae6d8ed6ea4720ee4fb621dd3328dc3ea2. The fix was made to the org.rhq.enterprise.server.measurement.MeasurementScheduleManagerBean class. Mass move to component = Monitoring Heiko, have you tested this in the perf env recently. If so, please mark this as verified, if not, please test. Mass-closure of verified bugs against JON. |