Perhaps you are seeing auto-snapshots before destructive events such as truncate or drop CF/KS.
On Mon, Aug 29, 2011 at 4:19 PM, Yan Chunlu <springri...@gmail.com> wrote: > just found the data dir consume a lot of space, which is because there was > many snapshots in it. > but I have set snapshot_before_compaction: false. is that possible that > cassandra create those snapshot automatically? could I delete them? > the dir names is strange(normally it should contain date info like this one: > 1309954201568-20110706snap): > 1309954201568-20110706snap 1313655860450 1313657278693 1313658563469 > 1313660368230 1313661946829 1313673041895 1313978414627 1313993893774 > 1313994151125 > 1309954218127-20110706snap 1313655977397 1313657539385 1313658794663 > 1313660540216 1313665575966 1313684621571 1313981129181 1313993899685 > 1313994436489 > 1309954367559-20110706snap 1313656318839 1313657769222 1313659117385 > 1313660791414 1313670545340 1313688842151 1313993527645 1313994045681 > 1313994537282 > 1313655524690 1313656750454 1313657987723 1313659841880 > 1313661186615 1313671377867 1313821800047 1313993601281 1313994093703 > 1313994548470 > 1313655632823 1313657058205 1313658288595 1313660165346 > 1313661433088 1313672861562 1313822718672 1313993882415 1313994138019 > 1313994791621 > -- Jonathan Ellis Project Chair, Apache Cassandra co-founder of DataStax, the source for professional Cassandra support http://www.datastax.com
