Hello, > > We have been trying out the new Solrcloud from trunk and were able to setup a multi-slice solr configuration. But when ever we have multiple slices/shards, the number of threads used by solr reaches more than 25k during indexing and hence runs out of memory/native threads. It occurs even if we have the slices in different physical machines under embedded and standalone zookeeper instances. Solr runs under the default jetty server in cloud mode with solrconfig from the example(not multicore) directory. We index documents using the csv request handler. It would be great if someone could give some insights into this issue. Please let me know if you need more details about our setup.
Thanks
