sureshanaparti commented on a change in pull request #4428:
URL: https://github.com/apache/cloudstack/pull/4428#discussion_r519641322
##########
File path:
engine/storage/src/main/java/org/apache/cloudstack/storage/endpoint/DefaultEndPointSelector.java
##########
@@ -128,9 +132,24 @@ protected EndPoint findEndPointInScope(Scope scope, String
sqlBase, Long poolId)
}
}
+ List<Long> dedicatedHosts = dedicatedResourceDao.listAllHosts();
+
// TODO: order by rand() is slow if there are lot of hosts
sbuilder.append(") t where t.value<>'true' or t.value is null");
//Added for exclude cluster's subquery
- sbuilder.append(" ORDER by rand() limit 1");
+ sbuilder.append(" ORDER by ");
+ if (dedicatedHosts.size() > 0) {
+ Collections.shuffle(dedicatedHosts); // Randomize dedicated hosts
as well.
+ sbuilder.append("field(t.id, ");
+ Iterator<Long> hostIt = dedicatedHosts.iterator();
+ while (hostIt.hasNext()) { // put dedicated hosts at the end of
the result set
Review comment:
> @sureshanaparti I ran the SQL multiple times, the dedicated hosts were
always at the end of the result set.
Let's consider host ids: 3, 4 are dedicated, then the query does " _ORDER by
field(t.id, '3','4'), rand() limit 1_". Am I correct?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
