[infinispan-issues] [JBoss JIRA] (ISPN-2139) ping_on_startup ignored
[
https://issues.jboss.org/browse/ISPN-2139?page=com.atlassian.jira.plugin....
]
Galder Zamarreño commented on ISPN-2139:
----------------------------------------
@Dan, @Michal, thx for the clarifications.
There's two ways to fix this:
1. Make ping be a retry-on-failure operation, but that's not really the intention of
the operation. The idea of ping is to check whether a particular server is up or down. As
Michal said, we're reusing ping to check whether a cache name is defined in the remote
server too.
2. Make the cache name in getCache() be controllable by ping_on_startup. This sounds more
sensible.
ping_on_startup ignored
-----------------------
Key: ISPN-2139
URL: https://issues.jboss.org/browse/ISPN-2139
Project: Infinispan
Issue Type: Bug
Affects Versions: 5.1.5.FINAL
Reporter: Michal Linhard
Assignee: Galder Zamarreño
Labels: jdg6
Fix For: 5.1.x, 5.2.0.ALPHA2, 5.2.0.FINAL
When a configuration property
infinispan.client.hotrod.ping_on_startup=false
is passed to the properties of RemoteCacheManager, this
avoids ping in RemoteCacheManager.start() but
RemoteCacheManager.getCache() does ping without checking this property
and thus rendering this setting useless for scenario
where we want to specify
infinispan.client.hotrod.server_list
with list of servers with some of the servers possibly not available yet.
In such case we would expect the client to failover to first working server
and retrying the first operation we want to do with the cache.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira