[ 
https://issues.apache.org/jira/browse/SOLR-10604?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15995681#comment-15995681
 ] 

Mark Miller commented on SOLR-10604:
------------------------------------

I have some memory of filing an issue for this or something like this before. I 
think at the time the tests did not beast fully though and so I didn't want to 
add any new complications before fixing the current issues. AFAIK, they are in 
a lot better shape now, but we would want to beast well before enabling this 
stuff.

> ChaosMonkey doesn't expire session or cause connection losses by default
> ------------------------------------------------------------------------
>
>                 Key: SOLR-10604
>                 URL: https://issues.apache.org/jira/browse/SOLR-10604
>             Project: Solr
>          Issue Type: Bug
>      Security Level: Public(Default Security Level. Issues are Public) 
>          Components: Tests
>            Reporter: Tomás Fernández Löbbe
>
> It looks like the intention in ChaosMonkey is that, unless explicitly set, it 
> should randomly cause session expiration and connection loss:
> {code:java}
>     if (EXP != null) {
>       expireSessions = EXP; 
>     } else {
>       expireSessions = chaosRandom.nextBoolean();
>     }
>     if (CONN_LOSS != null) {
>       causeConnectionLoss = CONN_LOSS;
>     } else {
>       causeConnectionLoss = chaosRandom.nextBoolean();
>     }
> {code}
> but {{EXP}} and {{CONN_LOSS}} are defined as:
> {code:java}
>   private static final Boolean CONN_LOSS = 
> Boolean.valueOf(System.getProperty("solr.tests.cloud.cm.connloss", null));
> private static final Boolean EXP = 
> Boolean.valueOf(System.getProperty("solr.tests.cloud.cm.exp", null));
> {code}
> Which makes them "false" (not null) unless those properties are set. 
> The fix is trivial, but we should see how it affects our tests before 
> submitting



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to