Tags
People
Badges
Help-
CORE FRAMEWORK
FEATURED COMPONENTS
-
RESOURCES
-
LEARN ZK
RESOURCES
Hello, in prod env (zk7, tomcat8) we have such situation https://gyazo.com/18b70ad2854a7f13cbe47242db19dc26 Is it normal? What can be a reason, event-thread? config
<system-config>
<disable-event-thread>false</disable-event-thread>
<ui-factory-class>org.zkoss.zk.ui.http.SerializableUiFactory
</ui-factory-class>
<disable-zscript>false</disable-zscript>
</system-config>
Can u contact for this issue?
Hey there,
Looking at the screenshot:
-894 active sessions: Is that the relative expected number of web browsers reaching your service during the session timeout period? (last 30mn I would guess)
-Session type is serializable session, which makes sense since you are using the serializable UI factory. Not a big difference, just implementing the serialize / deserialize workflows.
-creation time: 2020-06-08 (I'm assuming that this screenshot was taken around that time? If the screenshot was taken recently, then I'd expect the session would have been cleaned up by now.)
-other attributes: Internal request count (concurent requests, expected 0 unless user is currently requesting stuff from multiple servlets / portlets in the same session) / PI (timeout timer), seem normal at first glance.
This said, you might still get some hanging threads in there. Clustering (serialization) and event threads are not supposed to be used a the same time. (source here)
You might want to raise a ticket with support directly if you encounter a specific issue.
Asked: 2020-07-17 15:39:14 +0800
Seen: 8 times
Last updated: Jul 20 '20
