I am going a different direction. I'll load the data to a single instance and then distribute a copy of the partition to each host in the Multi-Master pool. It seems to be a faster process and no errors have occurred going this route.
Thanks for thew help. On Sat, Jun 20, 2015 at 10:41 AM, Kiran Ayyagari <[email protected]> wrote: > On Sat, Jun 20, 2015 at 11:24 PM, Ezsra McDonald <[email protected] > > > wrote: > > > As I was saying before I prematurely hit the send key: > > > > I am new to ApacheDS M20. I attempted to import 350k entries to a > > Multi-Rplication setup using ldapadd last night and found this in the > logs > > this morning. I really don't know what it wants me to do. > > > > "check that your keys are immutable, and that you have used > synchronization > > > properly" > > > > > > > Where do I find this information at? I did check my replication setup and > > it looks right. > > > > This entry was repeated in the logs: > > > > INFO | jvm 1 | 2015/06/20 07:30:16 | > java.lang.IllegalStateException: > > > Entry.next=null, > > > > > > data[removeIndex]=6f9128a9-c235-4d90-aecb-7e9ecb58441a=java.lang.Object@62dbd1cc > > > previous=6f9128a9-c235-4d90-aecb-7e9ecb58441a=java.lang.Object@62dbd1cc > > > key=938ea455-e1e5-4eca-bf2a-d99a00500865 value=java.lang.Object@271f5be > > > size=1000 maxSize=1000 Please check that your keys are immutable, and > > that > > > you have used synchronization properly. If so, then please report this > to > > > [email protected] as a bug. > > > it is properly synchronized, so I don't see why it is complaining, never > the less I have to get > rid of this LRUMap, it is not the first time we had issues with it (we had > a serious problem with in Mavibot) > > > > INFO | jvm 1 | 2015/06/20 07:30:16 | at > > > org.apache.commons.collections.map.LRUMap.reuseMapping(LRUMap.java:301) > > > INFO | jvm 1 | 2015/06/20 07:30:16 | at > > > org.apache.commons.collections.map.LRUMap.addMapping(LRUMap.java:267) > > > INFO | jvm 1 | 2015/06/20 07:30:16 | at > > > > > > org.apache.commons.collections.map.AbstractHashedMap.put(AbstractHashedMap.java:284) > > > INFO | jvm 1 | 2015/06/20 07:30:16 | at > > > > > > org.apache.directory.server.ldap.replication.consumer.ReplicationConsumerImpl.getLockFor(ReplicationConsumerImpl.java:1406) > > > INFO | jvm 1 | 2015/06/20 07:30:16 | at > > > > > > org.apache.directory.server.ldap.replication.consumer.ReplicationConsumerImpl.handleSearchResultEntry(ReplicationConsumerImpl.java:356) > > > INFO | jvm 1 | 2015/06/20 07:30:16 | at > > > > > > org.apache.directory.server.ldap.replication.consumer.ReplicationConsumerImpl.doSyncSearch(ReplicationConsumerImpl.java:769) > > > INFO | jvm 1 | 2015/06/20 07:30:16 | at > > > > > > org.apache.directory.server.ldap.replication.consumer.ReplicationConsumerImpl.startSync(ReplicationConsumerImpl.java:566) > > > INFO | jvm 1 | 2015/06/20 07:30:16 | at > > > org.apache.directory.server.ldap.LdapServer$2.run(LdapServer.java:743) > > > INFO | jvm 1 | 2015/06/20 07:30:16 | at > > > java.lang.Thread.run(Thread.java:745) > > > > > > > > > Also, The following was repeated in the logs for the system the ldapadd > > targeted. The message repeated for what may be every entry are the > > following. The database was empty to begin with. > > > > INFO | jvm 1 | 2015/06/19 19:20:54 | [19:20:54] WARN > > > [org.apache.directory.api.ldap.model.entry.DefaultAttribute] - > > > ERR_04486_VALUE_ALREADY_EXISTS The value 'organizationalPerson' already > > > exists in the attribute (objectClass) > > > > > > this can be ignore, just a log but at wrong level > > > > > Many hours later the logs for the targeted import node logged the > > following. It then became unresponsive. > > > > NFO | jvm 1 | 2015/06/20 07:30:20 | [07:30:20] WARN > > > [org.apache.directory.server.ldap.LdapProtocolHandler] - Unexpected > > > exception forcing session to close: sending disconnect notice to > client. > > > INFO | jvm 1 | 2015/06/20 07:30:20 | java.lang.OutOfMemoryError: > GC > > > overhead limit exceeded > > > > > > > can you take a memory dump of this process, that would help me debug it > better > > > > > There are four nodes in the Multi-Master configuration. Two in each data > > center. The contextCSN does not match on the node that reported the > > java.lang.IllegalStateException. > > > can you share the configuration files of all the nodes (strip the passwords > and host names) > > > > -- > Kiran Ayyagari > http://keydap.com >
