Delete your recovery log tables,
Alex wrote: > Alex wrote: >> Alex wrote: >>> Alex wrote: >>>> When I attempt to start the first controller of a pair (in a >>>> horizontal scalability setup), I get "not the last man down" error: >>>> >>>> ==========8<============= >>>> 14:49:53,335 INFO controller.virtualdatabase.myDB Recovery log >>>> size: 0 >>>> 14:49:53,336 INFO controller.virtualdatabase.myDB Using Hedera >>>> properties file: /hedera_jgroups.properties >>>> >>>> ------------------------------------------------------- >>>> GMS: address is 192.168.0.128:32792 >>>> ------------------------------------------------------- >>>> 14:49:58,134 INFO controller.virtualdatabase.myDB Group myDB >>>> connected to Member(address=/192.168.0.128:32792, uid=myDB) >>>> 14:49:58,134 INFO controller.virtualdatabase.myDB First controller >>>> in group myDB >>>> 14:49:59,172 ERROR controller.xml.DatabasesParser The virtual >>>> database myDB could not be added to the controller (because of >>>> Error while joining group myDB >>>> (org.continuent.sequoia.common.exceptions.VirtualDatabaseException: >>>> NOT STARTING VDB (not the last man down).)) >>>> java.lang.Exception: Error while joining group myDB >>>> (org.continuent.sequoia.common.exceptions.VirtualDatabaseException: >>>> NOT STARTING VDB (not the last man down).) >>>> at >>>> org.continuent.sequoia.controller.virtualdatabase.DistributedVirtualDatabase.joinGroup(DistributedVirtualDatabase.java:767) >>>> >>>> at >>>> org.continuent.sequoia.controller.xml.DatabasesParser.endElement(DatabasesParser.java:702) >>>> >>>> ==========8<============= >>>> >>>> I'm not sure exactly what it's saying here. Is this problem caused >>>> by having the virtual database auto-loading? (Because it tries >>>> loading with only one controller there, presumably, if that's the >>>> problem...) >>> Btw, I've changed the "autoEnableBackends" option to false in my >>> controller.xml, and still no joy, so I don't think that is the >>> issue.... >> Would also like to add that there aren't any other controllers still >> running elsewhere, which is what the error made me think of! >> Alex > Just to add yet more: > I've come to the understanding that the error indicates that the > controller is not the last one shutdown (by which it means "disable" > command I presume?), and that one should start controllers in the > reverse order that they were shut down. > Thing is, how do I fix the current muddle? The controllers on both > machines report they are "not the last man down" when I try them. Do I > have to reset some state somewhere or delete a file somewhere? > > ta, > Alex > > > > > > _______________________________________________ > Sequoia mailing list > [email protected] > https://forge.continuent.org/mailman/listinfo/sequoia >
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________ Sequoia mailing list [email protected] https://forge.continuent.org/mailman/listinfo/sequoia
