How to rebalance a cluster?

2014-03-07 Thread Oleg Dulin
I have the following situation: 10.194.2.5RAC1Up Normal 378.6 GB50.00% 0 10.194.2.4RAC1Up Normal 427.5 GB50.00% 127605887595351923798765477786913079295 10.194.2.7RAC1Up Normal 350.63 GB 50.00%

Weird timeouts

2014-03-07 Thread Joel Samuelsson
I try to fetch all the row keys from a column family (there should only be a couple of hundred in that CF) in several different ways but I get timeouts whichever way I try: Through the cassandra cli: Fetching 45 rows is fine: list cf limit 46 columns 0; . . . 45 Rows Returned. Elapsed time: 298 ms

replication_factor: ?

2014-03-07 Thread Daniel Curry
I would like to know on what is the rule of thumb for "replication_factor:" number? I think the answer is depends on how many nodes one has? IE: three nodes will be the number 3. What would happen it I put the number 2 for a three node cluster? We are using both 3.2.4 and 3.1.3 ( that will

Re: Backup/Restore in Cassandra

2014-03-07 Thread Jonathan Lacefield
Hello, Full snapshot forces a flush, yes. Incremental hard-links to SSTables, yes. This question really depends on how your cluster was "lost". Node Loss: You would be able to restore a node based on restoring backups + commit log or just by using repair. Cluster Loss: (all nod

Re: replication_factor: ?

2014-03-07 Thread John Pyeatt
You really don't want to set your RF to the same value as the number of nodes in your cluster for a variety of reasons. The biggest one being that if you have a node go down, your entire database is essentially down because you will be unable to fulfil any requests because the RF can never be met.

remove

2014-03-07 Thread Grga Pitich

Re: replication_factor: ?

2014-03-07 Thread Jonathan Lacefield
Hello, The rule of thumb depends on your use case, particularly your consistency requirements. Typical configuration is to leverage RF3. Here's documentation on consistency levels: http://www.datastax.com/documentation/cassandra/1.2/cassandra/dml/dml_config_consistency_c.html If you had a 3

Re: replication_factor: ?

2014-03-07 Thread Daniel Curry
Thank you. Your answer makes sense, Is this documented any where On 03/07/2014 07:37 AM, John Pyeatt wrote: You really don't want to set your RF to the same value as the number of nodes in your cluster for a variety of reasons. The biggest one being that if you have a node go down, your entire

Re: replication_factor: ?

2014-03-07 Thread Daniel Curry
Thank you for the link, On 03/07/2014 07:40 AM, Jonathan Lacefield wrote: Hello, The rule of thumb depends on your use case, particularly your consistency requirements. Typical configuration is to leverage RF3. Here's documentation on consistency levels: http://www.datastax.com/documenta

Re: Weird timeouts

2014-03-07 Thread Ken Hancock
Are you on Cassandra 1.2 and can utilize the trace functionality? Might be an informative route. Ken On Fri, Mar 7, 2014 at 9:22 AM, Joel Samuelsson wrote: > I try to fetch all the row keys from a column family (there should only be > a couple of hundred in that CF) in several different ways

Re: Weird timeouts

2014-03-07 Thread Duncan Sands
Hi Joel, On 07/03/14 15:22, Joel Samuelsson wrote: I try to fetch all the row keys from a column family (there should only be a couple of hundred in that CF) in several different ways but I get timeouts whichever way I try: did you check the node logs for exceptions? You can get this kind of

RE: Backup/Restore in Cassandra

2014-03-07 Thread java8964
Hi, Jonathan: Thanks for your answer. My original goal of this question is not really related to backup/restore, but to see if we can skip the Full Snapshot during ETL transferring the data from SSTable files of Cassandra into another Hadoop Cluster. Right now, our production generates a full sn

Re: replication_factor: ?

2014-03-07 Thread Robert Coli
On Fri, Mar 7, 2014 at 7:26 AM, Daniel Curry wrote: > I would like to know on what is the rule of thumb for > "replication_factor:" number? > I think the answer is depends on how many nodes one has? IE: three nodes > will be the > number 3. What would happen it I put the number 2 for a three no

Re: How to rebalance a cluster?

2014-03-07 Thread Robert Coli
On Fri, Mar 7, 2014 at 6:00 AM, Oleg Dulin wrote: > I have the following situation: > > 10.194.2.5RAC1Up Normal 378.6 GB50.00% >0 > 10.194.2.4RAC1Up Normal 427.5 GB50.00% > 127605887595351923798765477786913079295 > 10.194.2.7RAC1

Driver documentation questions

2014-03-07 Thread Green, John M (HP Education)
I've been tinkering with both the C++ and Java drivers but in neither case have I got a good indication of how threading and resource mgmt should be implemented in a long-lived multi-threaded application server process.That is, what should be the scope of a builder, a cluster, session, and s

Re: Driver documentation questions

2014-03-07 Thread Peter Lin
you would create a new session. Don't create a new cluster, that will quickly exhaust the connections to the servers On Fri, Mar 7, 2014 at 3:42 PM, Green, John M (HP Education) < john.gr...@hp.com> wrote: > I've been tinkering with both the C++ and Java drivers but in neither > case have I got

Re: Compaction does not remove tombstones if column has higher TTL

2014-03-07 Thread Ken Hancock
I agree, that's totally unintuitive. I would have the same expectations that compaction is done on a row/column pair instead of simply at the row level. On Fri, Feb 28, 2014 at 11:44 AM, Keith Wright wrote: > FYI - I recently filed > https://issues.apache.org/jira/browse/CASSANDRA-6654 and wan

RE: replication_factor: ?

2014-03-07 Thread Donald Smith
Robert, please elaborate why you say "To make best use of Cassandra, my minimum recommendation is usually RF=3, N=6." I surmise that with any less than 6 nodes, you'd likely perform better with a sequential/single-node solution. You need at least six nodes to overcome the overheads from concur

read one -- internal behavior

2014-03-07 Thread James Lyons
I'm wondering about the following scenario. Consider a cluster of nodes with replication say 3. When performing a read at "read one" consistency and lets say my client isn't smart enough to route the request to the Cassandra node housing the data at first. the contacted node acts as a coordinator

Re: read one -- internal behavior

2014-03-07 Thread Jonathan Lacefield
B is the answer > On Mar 7, 2014, at 7:35 PM, James Lyons wrote: > > I'm wondering about the following scenario. > > Consider a cluster of nodes with replication say 3. > When performing a read at "read one" consistency and lets say my client isn't > smart enough to route the request to the Cass

Re: read one -- internal behavior

2014-03-07 Thread Russell Hatch
If you are using cqlsh, you can get a look at what's happening behind the scenes by enabling tracing with 'tracing on;' before executing a query. In this scenario you'll see 'Sending message to [ip address]' for each of the replicas. On Fri, Mar 7, 2014 at 5:44 PM, Jonathan Lacefield wrote: > B

Re: read one -- internal behavior

2014-03-07 Thread Jonathan Lacefield
Yikes my apologies. B is not the answer On Mar 7, 2014, at 8:24 PM, Russell Hatch wrote: If you are using cqlsh, you can get a look at what's happening behind the scenes by enabling tracing with 'tracing on;' before executing a query. In this scenario you'll see 'Sending message to [ip address]

Re: read one -- internal behavior

2014-03-07 Thread Michael Shuler
Actually, this is tunable in >= 2.0.2 ;) http://www.datastax.com/dev/blog/rapid-read-protection-in-cassandra-2-0-2 Michael On 03/07/2014 07:33 PM, Jonathan Lacefield wrote: Yikes my apologies. B is not the answer On Mar 7, 2014, at 8:24 PM, Russell Hatch mailto:rha...@datastax.com>> wrote: