Re: new nodetool ring output and unbalanced ring?

2012-09-10 Thread Guy Incognito
out of interest, why -100 and not -1 or + 1? any particular reason? On 06/09/2012 19:17, Tyler Hobbs wrote: To minimize the impact on the cluster, I would bootstrap a new 1d node at (42535295865117307932921825928971026432 - 100), then decommission the 1c node at

replace_token code?

2012-09-10 Thread Yang
it looks that by specifying replace_token, the old owner is not removed from gossip (which I had thought it would do). Then it's understandable that the old owner would resurface later and we get some warning saying that the same token is owned by both. I ran an example with a 2-node cluster,

Long-life TTL and extending TTL

2012-09-10 Thread Robin Verlangen
Hi there, I'm working on a project that might want to set TTL to roughly 7 years. However it might occur that the TTL should be reduced or extended. Is there any way of updating the TTL without being in need of rewriting the data back again? This would cause way to much overhead for this. If

Re: Long-life TTL and extending TTL

2012-09-10 Thread Oleg Dulin
You should create an index where you store references to your records. You can use composite column names where column name=composite(timestamp,key) then you would get a slice of all columns where timestamp part of the composite is = TTL in the past, and then iterate through them and delete

Re: Long-life TTL and extending TTL

2012-09-10 Thread Robin Verlangen
OK, that's an option indeed. However due to the amount of records it would also involve bucketing which makes it not the simplest option. Further more, there are lots of manual indexes referring to the keys of the actual events: all those indexes should also be updated. Best regards, Robin

Re: High commit size

2012-09-10 Thread Oleg Dulin
It is memory-mapped I/O. I wouldn't worry about it. BTW, Windows might not be the best choice to run Cassandra on. My experience running Cassandra on Windows has not been positive one. We no longer support Windows as our production platform. Regards, Oleg On 2012-09-10 09:00:02 +, Rene

Re: High commit size

2012-09-10 Thread Rene Kochen
The problem is that the system just freezes and nodes are dying. The system becomes very unresponsive and it always happens when the shareable amount of RAM reaches the total number of bytes in the system. Is there something in Windows that I can tune in order to avoid this behavior? I cannot

Re: Long-life TTL and extending TTL

2012-09-10 Thread Sylvain Lebresne
Is there any way of updating the TTL without being in need of rewriting the data back again? No, there isn't. If not, is running a Map/Reduce task on the whole data set the best option If the TTL is made rather infrequently and on a large percentage of the data, which seems to be your

RE: High commit size

2012-09-10 Thread Viktor Jevdokimov
Do not use mmap/auto on Windows, standard access mode only. In cassandra.yaml: disk_access_mode: standard Best regards / Pagarbiai Viktor Jevdokimov Senior Developer Email: viktor.jevdoki...@adform.commailto:viktor.jevdoki...@adform.com Phone: +370 5 212 3063, Fax +370 5 261 0453 J. Jasinskio

Re: High commit size

2012-09-10 Thread Rene Kochen
For performance reasons I switched to memory mapped IO. Is there a way to make memory-mapped IO work in Windows? Thanks! 2012/9/10 Viktor Jevdokimov viktor.jevdoki...@adform.com Do not use mmap/auto on Windows, standard access mode only. In cassandra.yaml: disk_access_mode:

Suggestions for tuning and monitoring.

2012-09-10 Thread Henrik Schröder
Hi all, We're running a small Cassandra cluster (v1.0.10) serving data to our web application, and as our traffic grows, we're starting to see some weird issues. The biggest of these is that sometimes, a single node becomes unresponsive. It's impossible to start new connections, or impossible to

RE: High commit size

2012-09-10 Thread Viktor Jevdokimov
We used Cassandra on Windows for more than a year in production for RTB and other staff, that requires lowest possible latency. We used mmap before issues like yours, switched to mmap index only and finally to standard. No big difference in performance, standard was most stable. The huge

Re: High commit size

2012-09-10 Thread Henrik Schröder
When we ran Cassandra on windows, we got better performance without memory mapped IO. We had the same problems your are describing, what happens is that Windows is rather aggressive about swapping out memory when all the memory is used, and it starts swapping out unused parts of the heap, which

any way to prefer just 3 column families for partial row caching

2012-09-10 Thread Hiller, Dean
We have 3 tables for all indexing we do called IntegerIndexing DecimalIndexing StringIndexing playOrm would prefer that only these rows are cached as every row in those tables are indices. Customers/Clients of playOrm tend to always hit the same index rows over and over as they are using the

Assertions running Cleanup on a 3-node cluster with Cassandra 1.1.4 and LCS

2012-09-10 Thread Rudolf van der Leeden
Hi, I'm getting 5 identical assertions while running 'nodetool cleanup' on a Cassandra 1.1.4 node with Load=104G and 80m keys. From system.log : ERROR [CompactionExecutor:576] 2012-09-10 11:25:50,265 AbstractCassandraDaemon.java (line 134) Exception in thread

Re: replace_token code?

2012-09-10 Thread Jim Cistaro
We have seen various issues from these replaced nodes hanging around. For clusters where a lot of nodes have been replaced, we see these replaced nodes having an impact on heap/GC and a lot of tcp timeouts/retransmits (because the old nodes no longer exist). As a result, we have begun

JVM 7, Cass 1.1.1 and G1 garbage collector

2012-09-10 Thread Oleg Dulin
I am currently profiling a Cassandra 1.1.1 set up using G1 and JVM 7. It is my feeble attempt to reduce Full GC pauses. Has anyone had any experience with this ? Anyone tried it ? -- Regards, Oleg Dulin NYC Java Big Data Engineer http://www.olegdulin.com/

Re: replace_token code?

2012-09-10 Thread Yang
Thanks Jim, looks I'll have to read into the code to understand what is happening under the hood yang On Mon, Sep 10, 2012 at 9:45 AM, Jim Cistaro jcist...@netflix.com wrote: We have seen various issues from these replaced nodes hanging around. For clusters where a lot of nodes have been

[RELEASE] Apache Cassandra 1.1.5 released

2012-09-10 Thread Sylvain Lebresne
The Cassandra team is pleased to announce the release of Apache Cassandra version 1.1.5. Cassandra is a highly scalable second-generation distributed database, bringing together Dynamo's fully distributed design and Bigtable's ColumnFamily-based data model. You can read more here:

Re: new nodetool ring output and unbalanced ring?

2012-09-10 Thread Tyler Hobbs
It leaves some breathing room for fixing mistakes, adding DCs, etc. The set of data in a 100 token range is basically the same as a 1 token range: nothing, statistically speaking. On Mon, Sep 10, 2012 at 2:21 AM, Guy Incognito dnd1...@gmail.com wrote: out of interest, why -100 and not -1 or +

Re: java.lang.NoClassDefFoundError when trying to do anything on one CF on one node

2012-09-10 Thread Thomas van Neerijnen
Hi Aaron Removing NodeInfo did do the trick, thanks. I see the ticket is already resolved, good news. Thanks for the help. On Fri, Sep 7, 2012 at 12:26 AM, aaron morton aa...@thelastpickle.comwrote: This is a problem… [default@system] list NodeIdInfo ; Using default limit of 100 ...

MySQL to Cassandra Migration tool

2012-09-10 Thread Morantus, James (PCLN-NW)
Hey folks, Can you recommend any tools to pull data from MySQL and pump it to Cassandra? Thanks in advanced. James

Re: MySQL to Cassandra Migration tool

2012-09-10 Thread Aaron Turner
On Mon, Sep 10, 2012 at 10:17 PM, Morantus, James (PCLN-NW) james.moran...@priceline.com wrote: Hey folks, Can you recommend any tools to pull data from MySQL and pump it to Cassandra? This: http://www.datastax.com/dev/blog/bulk-loading -- Aaron Turner http://synfin.net/ Twitter:

Re: [RELEASE] Apache Cassandra 1.1.5 released

2012-09-10 Thread Radim Kolar
this is first version from 1.1 branch i used in pre-production stress testing and i got lot of following errors: decorated key -1 != some number INFO [CompactionExecutor:10] 2012-09-11 02:22:13,586 CompactionController.java (line 172) Compacting large row

Re: Number of columns per row for Composite Primary Key CQL 3.0

2012-09-10 Thread aaron morton
In general wider rows take a bit longer to read, however different access patterns have different performance. I did some tests here http://www.slideshare.net/aaronmorton/cassandra-sf-2012-technical-deep-dive-query-performance and http://thelastpickle.com/2011/07/04/Cassandra-Query-Plans/ I

Re: Suggestions for tuning and monitoring.

2012-09-10 Thread aaron morton
It's impossible to start new connections, or impossible to send requests, or it just doesn't return anything when you've sent a request. If it's totally frozen it sounds like GC. How long does it freeze for? Despite that, we occasionally get OOM exceptions, and nodes crashing, maybe a few

Re: any way to prefer just 3 column families for partial row caching

2012-09-10 Thread aaron morton
playOrm would prefer that only these rows are cached as every row in those tables are indices CF level key and row caching is specified by the caching property. It can be either KEY, ROW or BOTH. So you can turn on the row cache, but only have some CF's use it. Cheers -

Re: Assertions running Cleanup on a 3-node cluster with Cassandra 1.1.4 and LCS

2012-09-10 Thread aaron morton
My question: Can these assertions be ignored? Or do I need to worry about it? That looks like a problem. Can you raise a ticket on https://issues.apache.org/jira/browse/CASSANDRA ? May be good to include information on: * how long you've been using Levelled Compaction. * Is this all CF's or

Re: JVM 7, Cass 1.1.1 and G1 garbage collector

2012-09-10 Thread Peter Schuller
I am currently profiling a Cassandra 1.1.1 set up using G1 and JVM 7. It is my feeble attempt to reduce Full GC pauses. Has anyone had any experience with this ? Anyone tried it ? Have tried; for some workloads it's looking promising. This is without key cache and row cache and with a pretty