[
https://issues.jboss.org/browse/ISPN-9033?page=com.atlassian.jira.plugin....
]
Sergey Chernolyas commented on ISPN-9033:
-----------------------------------------
Hi [~NadirX]
I understood it. But ... Can we optimize it?
For example, if cache has persistence without passivation ... we can ask cache store
about count of keys. It is fast.
I guess that we can divide the issue to several subtasks (one subtask is for each cache)
I have two caches with size more that 5 000 000 keys. All caches have cache store based on
RocksDB. And ... counting take one tread and count during 10-15 minutes. I can't open
Web interface for looking statistic. I can use CLI .... but it is too long to.
Web/CLI hangs if cache use cache store and has a lot of entries
---------------------------------------------------------------
Key: ISPN-9033
URL:
https://issues.jboss.org/browse/ISPN-9033
Project: Infinispan
Issue Type: Bug
Components: Core
Affects Versions: 9.2.1.Final
Reporter: Sergey Chernolyas
My cache configuration has RocksDB's cache. Cache has a more then 1 000 000 entries.
And it do usage of Web interfce and CLI impossible.
As I have investigated, problem can be in class
org.infinispan.commands.read.SizeCommand.
{code:java}
public Integer perform(InvocationContext ctx) throws Throwable {
long size = cache.keySet().stream().count();
if (size > Integer.MAX_VALUE) {
return Integer.MAX_VALUE;
} else {
return (int) size;
}
}
{code}
I think, the code must ask CacheStore for sizing.
--
This message was sent by Atlassian JIRA
(v7.5.0#75005)