Re: question on maximum disk seeks

Yes, I understand that. However, what I'm trying to understand is the internal structure of partition index. When a record associate with the same partition key is updated, we have two different records with different timestamps. There are chances of these two records being split across two

Re: question on maximum disk seeks

The partition index is never updated, as sstables are immutable. On Tue, Mar 21, 2017 at 9:40 AM preetika tyagi wrote: > Thank you Jan & Jeff for the responses. That was really useful. > > Jan - I have one follow-up question. When the data is spread over more > than one

ONE has much higher latency than LOCAL_ONE

I am seeing unexpected behavior: consistency level ONE increases read latency 99th percentile to ~108ms (95th percentile to 5ms-90ms) up from ~5ms (99th percentile) when using LOCAL_ONE. I am using DSE 5.0 with Datastax client 3.0.0. The client is configured with a TokenAwarePolicy wrapping a

Re: question on maximum disk seeks

Each sstable has it's own partition index, therefore it's never updated. On Tue, Mar 21, 2017 at 11:04 AM preetika tyagi wrote: > Yes, I understand that. However, what I'm trying to understand is the > internal structure of partition index. When a record associate with

How to add a node with zero downtime

Hi list, I'm using C* 3.10; authenticator: PasswordAuthenticator and authorizer: CassandraAuthorizer When adding a node and before |nodetool repair system_auth| finished all my clients die with: cassandra.cluster.NoHostAvailable: ('Unable to connect to any servers', {'10.100.100.19':

Re: question on maximum disk seeks

Oh I see. I understand it now. Thank you for the clarification! Preetika On Tue, Mar 21, 2017 at 11:07 AM, Jonathan Haddad wrote: > Each sstable has it's own partition index, therefore it's never updated. > > On Tue, Mar 21, 2017 at 11:04 AM preetika tyagi

Re: Issue with Cassandra consistency in results

The same issue is appearing in CQL Shell as well. 1) Entered into cqlsh 2) SET CONSISTENCY QUORUM; 3) Ran a select * with partition key in where cluase. First result gave 0 records, and Next records gave results. Its really freaking out us at the moment. And nothing in debug.log or system.log.

Re: How to add a node with zero downtime

Possible areas to check: - too few nodes (node overload) - you did not indicate either replication factor, number of nodes. Assume nodes are *rather* full. - network overload (check your TORS's errors, also the tcp stats on the relevant nodes) - look for stop the world garbage collection on

Re: Grouping time series data into blocks of times

Kairosdb (OpenTSDB clone forCassandra) is a TSDB that does this. MAy be you could have a look at it ? It has a daemon process that collects and groups data points into blobs before writing to cassandra. -- best, Alain

Re: ONE has much higher latency than LOCAL_ONE

Are you running a multi DC cluster? If yes do you have application in both data centers/regions ? On Tue, Mar 21, 2017 at 8:07 PM, Shannon Carey wrote: > I am seeing unexpected behavior: consistency level ONE increases read > latency 99th percentile to ~108ms (95th

Re: Issue with Cassandra consistency in results

Hi This issue used to appear with me . What I figured in my case was 1. I had 3 machines 2. Inserted the data with ONE consistency (i.e there is no guarantee that data was propagated to remaining nodes , cassandra is supposed to take care of that). 3. Later I figured also that one of

Re: ONE has much higher latency than LOCAL_ONE

On Wed, Mar 22, 2017 at 12:48 PM, Shannon Carey wrote: > > The cluster is in two DCs, and yes the client is deployed locally to each DC. First off, what is the goal of using ONE instead of LOCAL_ONE? If it's failover, this could be addressed with a RetryPolicy starting wth

Re: ONE has much higher latency than LOCAL_ONE

On Wed, Mar 22, 2017 at 1:11 PM, Nate McCall wrote: > > > On Wed, Mar 22, 2017 at 12:48 PM, Shannon Carey > wrote: > > > > The cluster is in two DCs, and yes the client is deployed locally to > each DC. > > First off, what is the goal of using ONE

Re: ONE has much higher latency than LOCAL_ONE

ONE means at least one replica node to ack the write, but doesn't require that the coordinator route the request to a node in the local data center. LOCAL_ONE was introduced to handle the case of when you have multiple data centers and cross data center traffic is not desirable. In multiple

Re: Scrubbing corrupted SStable.

The snapshots are hard links on the file system, so everything is included. You can use the "--no-snapshot" option to disable snapshots. On Tue, Mar 21, 2017 at 5:01 PM, Pranay akula wrote: > > I am trying to scrub a Column family using nodetool scrub, is it going to

Re: spikes in blocked native transport requests

See the details on: https://issues.apache.org/jira/browse/CASSANDRA-11363 You may need to add -Dcassandra.max_queued_native_transport_requests=4096 as a startup parameter. YMMV though, I suggest reading through the above to get a complete picture. On Mon, Mar 20, 2017 at 11:10 PM, Roland Otta

Re: ONE has much higher latency than LOCAL_ONE

The cluster is in two DCs, and yes the client is deployed locally to each DC. From: Matija Gobec > Reply-To: "user@cassandra.apache.org" > Date:

[Cassandra 3.0.9] Cannot allocate memory

Hi all, I am using Cassandra 3.0.9. while I am adding new server after some time I am getting below exception. JVM option is attaches. Hardware info: Ram 64 GB. Core: 40 Java HotSpot(TM) 64-Bit Server VM warning: INFO: os::commit_memory(0x7fe9c44ee000, 12288, 0) failed; error='Cannot

Re: question on maximum disk seeks

Thank you Jan & Jeff for the responses. That was really useful. Jan - I have one follow-up question. When the data is spread over more than one SSTable in case of updates as you mentioned, we will need two seeks per SSTable (one for partition index and another for SSTable itself). I'm curious to

Re: Consistency Level vs. Retry Policy when no local nodes are available

Thanks for the perspective Ben, it's food for thought. At minimum, it seems like the documentation should be updated to mention that the retry policy will not be consulted when using a local consistency level but with no local nodes available. That way, people won't be surprised by it. It