Re: Timeouts but returned consistency level is invalid
Jan is incorrect. Keyspaces do not have consistency levels set on them. Consistency Levels are always set by the client. You are almost certainly running into https://issues.apache.org/jira/browse/CASSANDRA-7947 which is fixed in 2.1.3 and 2.0.12. On Fri, Jan 30, 2015 at 8:37 AM, Michał Łowicki mlowi...@gmail.com wrote: Hi Jan, I'm using only one keyspace. Even if it defaults to ONE why sometimes ALL is returned? On Fri, Jan 30, 2015 at 2:28 PM, Jan cne...@yahoo.com wrote: HI Michal; The consistency level defaults to ONE for all write and read operations. However consistency level is also set for the keyspace. Could it be possible that your queries are spanning multiple keyspaces which bear different levels of consistency ? cheers Jan C* Architect On Friday, January 30, 2015 1:36 AM, Michał Łowicki mlowi...@gmail.com wrote: Hi, We're using C* 2.1.2, django-cassandra-engine which in turn uses cqlengine. LOCAL_QUROUM is set as default consistency level. From time to time we get timeouts while talking to the database but what is strange returned consistency level is not LOCAL_QUROUM: code=1200 [Coordinator node timed out waiting for replica nodes' responses] message=Operation timed out - received only 3 responses. info={'received_responses': 3, 'required_responses': 4, 'consistency': 'ALL'} code=1200 [Coordinator node timed out waiting for replica nodes' responses] message=Operation timed out - received only 1 responses. info={'received_responses': 1, 'required_responses': 2, 'consistency': 'LOCAL_QUORUM'} code=1100 [Coordinator node timed out waiting for replica nodes' responses] message=Operation timed out - received only 0 responses. info={'received_responses': 0, 'required_responses': 1, 'consistency': 'ONE'} Any idea why it might happen? -- BR, Michał Łowicki -- BR, Michał Łowicki
Re: Timeouts but returned consistency level is invalid
Thanks Philip. This explains why I see ALL. Any idea why sometimes ONE is returned? — Michał On Fri, Jan 30, 2015 at 4:18 PM, Philip Thompson philip.thomp...@datastax.com wrote: Jan is incorrect. Keyspaces do not have consistency levels set on them. Consistency Levels are always set by the client. You are almost certainly running into https://issues.apache.org/jira/browse/CASSANDRA-7947 which is fixed in 2.1.3 and 2.0.12. On Fri, Jan 30, 2015 at 8:37 AM, Michał Łowicki mlowi...@gmail.com wrote: Hi Jan, I'm using only one keyspace. Even if it defaults to ONE why sometimes ALL is returned? On Fri, Jan 30, 2015 at 2:28 PM, Jan cne...@yahoo.com wrote: HI Michal; The consistency level defaults to ONE for all write and read operations. However consistency level is also set for the keyspace. Could it be possible that your queries are spanning multiple keyspaces which bear different levels of consistency ? cheers Jan C* Architect On Friday, January 30, 2015 1:36 AM, Michał Łowicki mlowi...@gmail.com wrote: Hi, We're using C* 2.1.2, django-cassandra-engine which in turn uses cqlengine. LOCAL_QUROUM is set as default consistency level. From time to time we get timeouts while talking to the database but what is strange returned consistency level is not LOCAL_QUROUM: code=1200 [Coordinator node timed out waiting for replica nodes' responses] message=Operation timed out - received only 3 responses. info={'received_responses': 3, 'required_responses': 4, 'consistency': 'ALL'} code=1200 [Coordinator node timed out waiting for replica nodes' responses] message=Operation timed out - received only 1 responses. info={'received_responses': 1, 'required_responses': 2, 'consistency': 'LOCAL_QUORUM'} code=1100 [Coordinator node timed out waiting for replica nodes' responses] message=Operation timed out - received only 0 responses. info={'received_responses': 0, 'required_responses': 1, 'consistency': 'ONE'} Any idea why it might happen? -- BR, Michał Łowicki -- BR, Michał Łowicki
Re: Timeouts but returned consistency level is invalid
Hi Jan, I'm using only one keyspace. Even if it defaults to ONE why sometimes ALL is returned? On Fri, Jan 30, 2015 at 2:28 PM, Jan cne...@yahoo.com wrote: HI Michal; The consistency level defaults to ONE for all write and read operations. However consistency level is also set for the keyspace. Could it be possible that your queries are spanning multiple keyspaces which bear different levels of consistency ? cheers Jan C* Architect On Friday, January 30, 2015 1:36 AM, Michał Łowicki mlowi...@gmail.com wrote: Hi, We're using C* 2.1.2, django-cassandra-engine which in turn uses cqlengine. LOCAL_QUROUM is set as default consistency level. From time to time we get timeouts while talking to the database but what is strange returned consistency level is not LOCAL_QUROUM: code=1200 [Coordinator node timed out waiting for replica nodes' responses] message=Operation timed out - received only 3 responses. info={'received_responses': 3, 'required_responses': 4, 'consistency': 'ALL'} code=1200 [Coordinator node timed out waiting for replica nodes' responses] message=Operation timed out - received only 1 responses. info={'received_responses': 1, 'required_responses': 2, 'consistency': 'LOCAL_QUORUM'} code=1100 [Coordinator node timed out waiting for replica nodes' responses] message=Operation timed out - received only 0 responses. info={'received_responses': 0, 'required_responses': 1, 'consistency': 'ONE'} Any idea why it might happen? -- BR, Michał Łowicki -- BR, Michał Łowicki
Re: Timeouts but returned consistency level is invalid
HI Michal; The consistency level defaults to ONE for all write and read operations. However consistency level is also set for the keyspace. Could it be possible that your queries are spanning multiple keyspaces which bear different levels of consistency ? cheersJan C* Architect On Friday, January 30, 2015 1:36 AM, Michał Łowicki mlowi...@gmail.com wrote: Hi, We're using C* 2.1.2, django-cassandra-engine which in turn uses cqlengine. LOCAL_QUROUM is set as default consistency level. From time to time we get timeouts while talking to the database but what is strange returned consistency level is not LOCAL_QUROUM: code=1200 [Coordinator node timed out waiting for replica nodes' responses] message=Operation timed out - received only 3 responses. info={'received_responses': 3, 'required_responses': 4, 'consistency': 'ALL'} code=1200 [Coordinator node timed out waiting for replica nodes' responses] message=Operation timed out - received only 1 responses. info={'received_responses': 1, 'required_responses': 2, 'consistency': 'LOCAL_QUORUM'} code=1100 [Coordinator node timed out waiting for replica nodes' responses] message=Operation timed out - received only 0 responses. info={'received_responses': 0, 'required_responses': 1, 'consistency': 'ONE'} Any idea why it might happen? -- BR, Michał Łowicki