Re: Cassandra cross dc replication row isolationCassandra cross dc replication row isolation

2019-05-21 Thread Alexey Knyshev
Anyone? ср, 8 мая 2019 г. в 11:37, Alexey Knyshev : > Hi, thanks for your answers! > > > Are you asking if writes are atomic at the partition level? If so yes. > If you have N columns in a simple k/v schema, and you send a write with X/N > of those columns set, all X will be u

Cluster schema version choosing

2019-05-20 Thread Alexey Korolkov
Hello team, In some circumstances, my cluster was split onto two schema versions (half on one version, and rest on another) In the process of resolving this issue, I restarted some nodes. Eventually, nodes migrated to one schema, but it was not clear why they choose exactly this version of schema?

Re: Cassandra cross dc replication row isolationCassandra cross dc replication row isolation

2019-05-08 Thread Alexey Knyshev
m of nodes on the DC on >> which the Coordinator accepted the write. Similarly it would only check >> nodes in that DC. Quorum would check across DCs in the whole cluster. >> On May 7, 2019, 12:11 PM -0500, Alexey Knyshev , >> wrote: >> >> Hi there! >> >>

Re: Cassandra cross dc replication row isolationCassandra cross dc replication row isolation

2019-05-08 Thread Alexey Knyshev
m of nodes on the DC on >> which the Coordinator accepted the write. Similarly it would only check >> nodes in that DC. Quorum would check across DCs in the whole cluster. >> On May 7, 2019, 12:11 PM -0500, Alexey Knyshev , >> wrote: >> >> Hi there! >> >>

Cassandra cross dc replication row isolationCassandra cross dc replication row isolation

2019-05-07 Thread Alexey Knyshev
Hi there! Could someone please explain how Column Family would be replicated and "visible / readable" in the following scenario? Having multiple geo-distributed datacenters with significant latency (up to 100ms RTT). Let's name two of them A and B and consider the following 2 cases: 1. Cassand

Re[2]: Question about adding nodes to a cluster

2015-02-09 Thread Plotnik, Alexey
Sorry, No - you are not doing it wrong ^) Yes, Cassandra partitioner is based on hash ring. Doubling number of nodes is the best cluster exctending policy I've ever seen, because it's zero-overhead. Hashring - you get MD5 max (2^128-1), divide it by number of nodes (partitions) getting N point

Re: Question about adding nodes to a cluster

2015-02-09 Thread Plotnik, Alexey
Yes, Cassandra partitioner is based on hash ring. Doubling number of nodes is the best cluster exctending policy I've ever seen, because it's zero-overhead. Hashring - you get MD5 max (2^128-1), divide it by number of nodes (partitions) getting N points and then evenly distribute them across you

Re: Cassandra on Ceph

2015-01-31 Thread Plotnik, Alexey
1. What do you mean by "on top of Ceph"? 2. What's he goal? -- Original Message -- From: "Colin Taylor" mailto:colin.tay...@gmail.com>> To: "user@cassandra.apache.org" mailto:user@cassandra.apache.org>> Sent: 01.02.2015 12:26:42 Subject: Cassandra on Ceph I may be forced to run Cassandra

Re[2]: Test

2014-12-03 Thread Plotnik, Alexey
Alah Akbar -- Original Message -- From: "Servando Muñoz G." mailto:smg...@gmail.com>> To: "user@cassandra.apache.org" mailto:user@cassandra.apache.org>> Sent: 04.12.2014 16:12:32 Subject: RE: Test Saludos… Quien eres tu De: Castelain, Alain [mailto:alain.castel...@xerox.com

Re[2]: how wide can wide rows get?

2014-11-13 Thread Plotnik, Alexey
We have 380k of them in some of our rows and it's ok. -- Original Message -- From: "Hannu Kröger" mailto:hkro...@gmail.com>> To: "user@cassandra.apache.org" mailto:user@cassandra.apache.org>> Sent: 14.11.2014 16:13:49 Subject: Re: how wide can wide rows get? The theoretical limit is mayb

Re: Better option to load data to cassandra

2014-11-11 Thread Plotnik, Alexey
What have you tried? -- Original Message -- From: "srinivas rao" mailto:pinnakasrin...@gmail.com>> To: "Cassandra Users" mailto:user@cassandra.apache.org>> Sent: 11.11.2014 22:51:54 Subject: Better option to load data to cassandra Hi Team, Please suggest me the better options to load da

Re[2]: Redundancy inside a cassandra node

2014-11-08 Thread Plotnik, Alexey
doesn't affect customers. That is. -- Original Message -- From: "Jabbar Azam" mailto:aja...@gmail.com>> To: "user@cassandra.apache.org" mailto:user@cassandra.apache.org>> Sent: 08.11.2014 19:43:18 Subject: Re: Redundancy inside a cassandra node Hello Alexey, The

Re: Redundancy inside a cassandra node

2014-11-07 Thread Plotnik, Alexey
Cassandra is a cluster itself, it's not necessary to have redundant each node. Cassandra has replication for that. And also Cassandra is designed to run in multiple data center - am think that redundant policy is applicable for you. Only thing from your saying you can deploy is raid10, other don

Re: Increased Cassandra connection latency

2014-05-26 Thread Alexey
It looks like a memory leak to me. All Cassandra processes are with the following JVM settings configured: -Xms10240M -Xmx10240M -Xmn2400M. The OldGen reaches nearly the maximum and it looks like the oldgen objects not being garbage collected. Alexey On Fri, May 23, 2014 at 3:56 PM, Nate McCall

Increased Cassandra connection latency

2014-05-23 Thread Alexey
cies are alright again (average < 5ms, max 50ms). Any thoughts are greatly appreciated. Thanks, Alexey

Increased Cassandra connection latency

2014-05-23 Thread Alexey Sverdelov
cies are alright again (average < 5ms, max 50ms). Any thoughts are greatly appreciated. Thanks, Alexey

Increased Cassandra connection latency

2014-05-23 Thread Alexey Sverdelov
cies are alright again (average < 5ms, max 50ms). Any thoughts are greatly appreciated. Thanks, Alexey

RE: Tombstones

2014-05-18 Thread Plotnik, Alexey
It's located beside SSTables and has a name in format .json For `accounts` column family in `company` keyspace it looks like /company/accounts/accounts.json -Original Message- From: Dimetrio [mailto:dimet...@flysoft.ru] Sent: 18 мая 2014 г. 2:06 To: cassandra-u...@incubator.apache.org Su

750Gb compaction task

2014-03-12 Thread Plotnik, Alexey
After rebalance and cleanup I have leveled CF (SSTable size = 100MB) and a compaction Task that is going to process ~750GB: > root@da1-node1:~# nodetool compactionstats pending tasks: 10556 compaction typekeyspace column family completed total unit progr

RE: Multi-threaded sub-range repair

2014-02-23 Thread Plotnik, Alexey
UPD: Here we can see that repair process can be executed in parallel: http://www.datastax.com/dev/blog/advanced-repair-techniques From: Plotnik, Alexey Sent: 24 февраля 2014 г. 12:42 To: user@cassandra.apache.org Subject: Multi-threaded sub-range repair 1. How many parallel threads is

Multi-threaded sub-range repair

2014-02-23 Thread Plotnik, Alexey
1. How many parallel threads is safe to have for sub-range repair process running for a single node? 2. Is Repair process affected `concurrent_compactors` parameter? Should the `concurrent_compactors` meet the multithreaded repair process needs?

RE: Turn off compression (1.2.11)

2014-02-23 Thread Plotnik, Alexey
, Feb 18, 2014 at 2:58 PM, Plotnik, Alexey mailto:aplot...@rhonda.ru>> wrote: Compression buffers are located in Heap, I saw them in Heapdump. That is: == public class CompressedRandomAccessReader extends RandomAccessReader { ….. private ByteBuffer compressed; // <-

RE: Turn off compression (1.2.11)

2014-02-18 Thread Plotnik, Alexey
ent: 19 февраля 2014 г. 6:24 To: user@cassandra.apache.org Subject: Re: Turn off compression (1.2.11) On Mon, Feb 17, 2014 at 4:35 PM, Plotnik, Alexey mailto:aplot...@rhonda.ru>> wrote: As an aside, 1.2.0 beta moved a bunch of data related to compression off the heap. If you were to try t

RE: Turn off compression (1.2.11)

2014-02-18 Thread Plotnik, Alexey
, Alexey mailto:aplot...@rhonda.ru>> wrote: After analyzing Heap I saw this buffer has a size about 70KB per SSTable. I have more than 30K SSTables per node. I'm thinking your problem is not compression, it's using the old 5mb default for Level Compaction and having 30,000 SSTa

Turn off compression (1.2.11)

2014-02-17 Thread Plotnik, Alexey
Each compressed SSTable uses additional transfer buffer in CompressedRandomAccessReader instance. After analyzing Heap I saw this buffer has a size about 70KB per SSTable. I have more than 30K SSTables per node. I want to turn off a compression for this column family to save some Heap. How can I

Hadoop kills Cassandra

2014-01-27 Thread Plotnik, Alexey
For some reason when my map-reduce job is almost complete, all mappers (~40) begin to connect to a single Cassandra node. This noe then die due to Java Heap space error. It looks like Hadoop is misconfigured. Valid behavior for me: Each mapper should iterate only a local node. How can I configu

RE: one big cluster vs multiple smaller clusters

2013-10-13 Thread Plotnik, Alexey
If you are talking about scaling: Cassandra scaling is absolutely horizontal without namenodes or other Mongo-bulshit-like intermediate daemons. And that’s why one big cluster has the same throughput as many smaller clusters. What will you do when your small clusters will exceed it’s capacity? Ca

Re: unnecessary tombstone's transmission during repair process

2012-10-19 Thread Alexey Zotov
Gus, we've found the cause. It was a problem in Cassandra, but it has been already fixed in cassandra 1.1.6. Commit with the problem: 2c69e2ea757be9492a095aa22b5d51234c4b4102 You can see it at https://issues.apache.org/jira/secure/attachment/12544204/CASSANDRA-4561-CS.patch Commit with the fix:

Re: Cassandra nodes loaded unequally

2012-10-12 Thread Alexey Zotov
es,keyspace=,columnfamily=","WriteCount" Also I suggest to check output of "nodetool compactionstats". -- Alexey

Re: unnecessary tombstone's transmission during repair process

2012-10-12 Thread Alexey Zotov
xception e) +{ +throw new RuntimeException(e); +} +digest.update(buffer.getData(), 0, buffer.getLength()); +} + +@Override public long getMarkedForDeleteAt() { return timestamp; -- Best regards** Zotov Alexey Grid Dynamics Skype: azotcsit

unnecessary tombstone's transmission during repair process

2012-10-11 Thread Alexey Zotov
Hi Guys, I have a question about merkle tree construction and repair process. When mercle tree is constructing it calculates hashes. For DeletedColumn it calculates hash using value. Value of DeletedColumn is a serialized local deletion time. We know that local deletion time can be different on di

cassandra 1.2 beta in production

2012-10-10 Thread Alexey Zotov
Hi Guys, What known critical bugs are there that couldn't allow to use 1.2 beta 1 in production? We don't use cql and secondary indexes. -- Best regards** Zotov Alexey Grid Dynamics Skype: azotcsit