Re: Recreating materialized views in cassandra

IRA describing: what's the workload/queries and how does it end up in an inconsistent state if you can reproduce it? On Wed, 29 Jul 2020 at 20:49, Jasonstack Zhao Yang < jasonstack.z...@gmail.com> wrote: > > The cluster started to crash when some partitions in MV crossed 1 GB > size at

Re: Recreating materialized views in cassandra

ns cluster instability while creating and deleting mv's > > The cluster started to crash when some partitions in MV crossed 1 GB size >> at few nodes, whereas in other nodes it is less than 50 MB. > > > Should we be worried about this? > > On Mon, Jul 27, 2020 at 10:18 PM Ja

Re: Recreating materialized views in cassandra

Hi, > We are facing data inconsistency issues between base tables and materialized views. do you run "nodetool repair" on both base and view regularly? > What are all the possible scenarios that we should be watching out for in a production environment? more cpu/io/gc for populating views. >

Re: Materialized View's additional PrimaryKey column

Hi Jon, Do you have any clue what's the cause of downtime using MV? eg. memory pressure, or overloaded by view writes? Thanks. On Fri, 26 Jul 2019 at 13:59, mehmet bursali wrote: > Thank you again for Clear information Jon! i give up 珞 > > Android’de Yahoo Postadan gönderildi >

Re: Commit Log question

Hi Charulata, IMO, 64MB is fine unless you archive commit log or scan it for backup. Zhao Charulata Sharma (charshar) 于2017年4月28日周五 上午8:01写道： > Hi , > > Can anyone please tell me the implication of increasing the > commitlog_segment_size_in_mb > from the default value

Re: Drop tables takes too long

, then flush all schema info into sstable, then reads all on disk schema into memory (5k tables info + related column info).. > You also might need to increase the node count if you're resource constrained. More nodes won't help and most probably make it worse due to coordination. Zhao Yang On

Re: why dose it still have to seach in SSTable when getting data in memtable in the read flow？

Hi, Cassandra uses last-writetime-win strategy. In memory data doesn't mean it is the latest data due to custom write time, if data is also in Sstable, Cassandra has to read it and reconcile. Jasonstack On Mon, 27 Mar 2017 at 7:53 PM, 赵豫峰 wrote: > hello, I get the message

Re:Question about compaction strategy changes

hi Edwards, when changibg gc_grace_second, no compaction willbbe triggered. regards, jasonstack Sent from my Mi phoneOn Seth Edwards , Oct 22, 2016 11:37 AM wrote:Hello! We're using TWCS and we notice that if we make changes to the options to the window unit or size, it

Re: Does anyone store larger values in Cassandra E.g. 500 KB?

1. usually before storing object, serialization is needed, so we can know the size. 2. add "chunk id" as last clustering key. Vikas Jaiman 于2016年10月21日周五 下午11:46写道： > Thanks for your answer but I am just curious about: > > i)How do you identify the size of the object

Re: regarding drain process

Hi Varun, It looks like a scheduled job that runs "nodetool drain".. Zhao Yang Varun Barala <varunbaral...@gmail.com>于2016年9月25日周日 下午7:45写道： > Jeff Jirsa thanks for your reply!! > > We are not using any chef/puppet and It happens only at one node other > nodes

Re: Read data from specific node in cassandra

Hi, Can you check LoadBalancing Policy -> whiteList ? jasonstack Varun Barala 于2016年5月5日周四 下午5:40写道： > Hi Siddharth Verma, > > You can define consistency level LOCAL_ONE. > > and you can applyh consistency level during statement creation. > > like this ->

StatusLogger is logging too many information

Hi, Currently StatusLogger will log info when there are dropped messages or GC more than 200 ms. In my use case, there are about 1000 tables. The status-logger is logging too many information for each tables. I wonder is there a way to reduce this log? for example, only print the thread pool

Re: Cassandra table limitation

gle cluster to handle the different (probably competing) workloads >> effectively. >> >> Mike >> >> On Tue, Apr 5, 2016 at 8:40 PM, jason zhao yang < >> zhaoyangsingap...@gmail.com> wrote: >> >>> Hi Jack, >>> >>> Thanks

Re: Cassandra table limitation

les. > > On Mon, Apr 4, 2016 at 10:17 AM, jason zhao yang < > zhaoyangsingap...@gmail.com> wrote: > >> Hi, >> >> This is Jason. >> >> Currently, I am using C* 2.1.10, I want to ask what's the optimal number >> of tables I should creat

read time coprocessor?

This is Yang from Apache Kylin project. We are thinking about using Cassandra instead of HBase as storage. I searched and read around and still have one question. Does Cassandra support read time coprocessor that allows moving computation to data node before scan result is returned? This shall

Re: linearizable consistency / Paxos ?

), then any subsequent operation on this partition key will commit this stalled operation before starting its own. On Mon, Aug 3, 2015 at 4:30 AM, Yang tedd...@gmail.com wrote: this link http://www.datastax.com/dev/blog/lightweight-transactions-in-cassandra-2-0 talks about linearizable

Re: linearizable consistency / Paxos ?

because it explains briefly the semantics On Mon, Aug 3, 2015 at 11:32 AM, Yang tedd...@gmail.com wrote: thanks for your answer DuyHai. I understand Paxos. but I think your description seems missing one important point: in the example you gave, a series of ongoing operation (INSERT

linearizable consistency / Paxos ?

is the fundamental difference between the standard replication protocol and Paxos that prevents us from implementing a 2-pc on top of the standard protocol? Thanks! yang

Re: question about bootstrapping sequence

I'm wondering how the Cassandra protocol brings a newly bootstrapped node up to speed. for ease of illustration, let's say we just have one key, K, and the value is continually updated: 1,2 ,3 ,4 originally we have 1 node, A, now node B joins, and needs to bootstrap and get its newly

Re: question about bootstrapping sequence

are different. On Tue, Jul 28, 2015 at 12:40 PM, Robert Coli rc...@eventbrite.com wrote: On Tue, Jul 28, 2015 at 1:31 AM, Yang tedd...@gmail.com wrote: I'm wondering how the Cassandra protocol brings a newly bootstrapped node up to speed. Bootstrapping nodes get extra replicated copies

Re: question about bootstrapping sequence

at 1:01 PM, Yang tedd...@gmail.com wrote: Thanks. but I don't think having more nodes in the example changes the issue I outlined. say u have just key X, rf = 3, nodes A, B, D are responsible for X. in stable mode, the updates X=1, 2, 3, goes to all 3 servers. then at this time, node C

Re: Counters 2.1 Accuracy

that the new counter implementations are safer although I'm not sure what that means in practice. Will the counters be 99.99% accurate? How often will they be over or under counted? Thanks, Mike. -- Thanks, Phil Yang

Re: Tables showing up as our_table-147a2090ed4211e480153bc81e542ebd/ in data dir

as our_table-147a2090ed4211e480153bc81e542ebd/ instead of as our_table/ Why would that happen? We're also seeing lagging compactions and high cpu usage. Thanks, Don -- Thanks, Phil Yang

Re: Is 2.1.5 ready for upgrade?

find that there are a lot of patches in 2.1.5. Is it ready for upgrade? I personally would not run either version in production at this time, but if forced, would prefer 2.1.5 over 2.1.2. =Rob -- Cheers, Brian http://www.integrallis.com -- Thanks, Phil Yang

Re: Getting ParNew GC in ... CMS Old Gen ... in logs

-- Thanks, Phil Yang

Re: Re-bootstrap node after disk failure

, Phil Yang ud1...@gmail.com wrote: you can use nodetool rebuild in this node. 2015-03-25 9:20 GMT+08:00 Flavien Charlon flavien.char...@gmail.com: Hi, What is the process to re-bootstrap a node after hard drive failure (Cassandra 2.1.3)? This is the same node as previously, but the data

Re: Re-bootstrap node after disk failure

like to re-bootstrap it from the data stored on the other nodes of the cluster (I have RF=3). I am not using vnodes. Thanks Flavien -- Thanks, Phil Yang

Re: Steps to do after schema changes

column or modifying any table properties? Thanks Ajay -- Thanks, Phil Yang

Re: Node stuck in joining the ring

indexes. See CASSANDRA-8798 ~mck -- - Nate McCall Austin, TX @zznate Co-Founder Sr. Technical Consultant Apache Cassandra Consulting http://www.thelastpickle.com -- Thanks, Phil Yang

What are the factors that affect the release time of each minor version?

release more frequently? Or we may make a rule to decide if we need release a new version? For example: If the latest version was released two weeks ago, or after the latest version we have already resolved 20 issues, we should release a new minor version. -- Thanks, Phil Yang

Re: Counter Column

-- Thanks, Phil Yang

Re: Counter Column

sorry for typo.. timestamp which Cassandra uses is independent on the timezone. Usually, it is recommended to use NTP to reduce the difference of timestamps in each nodes 2014-12-27 21:20 GMT+08:00 Phil Yang ud1...@gmail.com: In java, http://docs.oracle.com/javase/7/docs/api/java/lang

Re: 2014 nosql benchmark

Today I've also seen this benchmark in Chinese websites. SequoiaDB seems come from a Chinese startup company, and in db-engines ranking http://db-engines.com/en/ranking it's score is 0.00. So IMO I have to say I think this benchmark is a soft sell. They compare three databases, two written by c++

Re: full gc too often

testing result for the relationship between the pressure and the safety heap size? We know query a slice with many tombstones is not a good use case, but query a slice without tombstones may be a common use case, right? On Thu Dec 04 2014 at 8:43:38 PM Philo Yang ud1...@gmail.com wrote: I have

full gc too often

? Is it because there are too many objects in heap can't be recycled? I think review the table scheme designing and add new nodes into cluster is a good idea, but I still want to know if there is any other reason causing this trouble. Thanks, Philo Yang

Re: full gc too often

Yang 2014-12-05 12:34 GMT+08:00 Tim Heckman t...@pagerduty.com: On Dec 4, 2014 8:14 PM, Philo Yang ud1...@gmail.com wrote: Hi,all I have a cluster on C* 2.1.1 and jdk 1.7_u51. I have a trouble with full gc that sometime there may be one or two nodes full gc more than one time per minute

random thought: rewrite Cassandra using Scala ?

thought: would there be any value to rewrite the code in Scala? thanks Yang

RangeTombstoneList.diff AssertionError in 2.1.0

(SEPWorker.java:103) [apache-cassandra-2.1.0.jar:2.1.0] at java.lang.Thread.run(Thread.java:744) [na:1.7.0_51] I have also posted this issue on JIRA https://issues.apache.org/jira/browse/CASSANDRA-8013 Thanks, Philo Yang

Re: why bloom filter is only for row key?

Thanks Rob Thanks, Philo Yang 2014-09-16 2:12 GMT+08:00 DuyHai Doan doanduy...@gmail.com: Nice catch Rob On Mon, Sep 15, 2014 at 8:04 PM, Robert Coli rc...@eventbrite.com wrote: On Sun, Sep 14, 2014 at 11:22 AM, Philo Yang ud1...@gmail.com wrote: After reading some docs, I find

Re: why bloom filter is only for row key?

filter's property between row keys and row keys+column names automatically or by user's config? Thanks, Philo Yang 2014-09-15 2:45 GMT+08:00 DuyHai Doan doanduy...@gmail.com: Hello Philo Building bloom filter for column names (what you call column key) is technically possible but very

why bloom filter is only for row key?

on? Thanks, Philo Yang

Re: Best practices for frequently updated columns

) to cache the read request. Write into C* and redis both and read only from redis for the latest bar. Read from C* for previous bars if the memory size is bottleneck. Thanks, Philo Yang 2014-08-14 2:13 GMT+08:00 Robert Coli rc...@eventbrite.com: On Wed, Aug 13, 2014 at 8:01 AM, Jeremy Jongsma

select many rows one time or select many times?

user='someuser' and word = 'b'; select from word where user='someuser' and word = 'c'; . Which method will cause lower pressure on Cassandra cluster? Thanks, Philo Yang

Re: ghost table is breaking compactions and won't go away… even during a drop.

... Thanks, Philo Yang 2014-07-17 10:32 GMT+08:00 Kevin Burton bur...@spinn3r.com: you rock… glad it's fixed in 2.1… :) On Wed, Jul 16, 2014 at 7:05 PM, graham sanderson gra...@vast.com wrote: Known issue deleting and recreating a CF with the same name, fixed in 2.1 (manifests in lots of ways

Is timestamp a reserved word in CQL?

a reserved word in CQL? I'm using Cassandra 2.0.6. Thanks, Philo Yang

How to deal with SSTable FileNotFoundException?

) I think the reason is I have dropped this table and recreated it later. And I find this Exception blocks the stream from these nodes to others so bootstrap or rebuild can't finish and is stuck in 100% forever. What should I do to refresh the SSTable list in these nodes? -- Regards, Zhe Yang

Node never know a table has been DROP or CREATE if its gossip is disabled while executing this query

knows there is a table but I drop it when one node's gossip is disabled, this node will never know the table has been dropped. Is this a bug? -- Regards, Zhe Yang

Re: Cassandra 1.20 with Cloudera Hadoop (CDH4) Compatibility Issue

of the old mapred API. I don't see the C* project back porting their code at this time and if anything Cloudera should update their release!! On Feb 15, 2013, at 9:48 PM, Yang Song xfil...@gmail.com wrote: It is interesting though. I am using CDH4 which contains hadoop 0.20, and I am using

Re: Starting Cassandra

) ... 17 more I did not install Hadoop on this cluster but apparently it wants to use it. Should I first build a Hadoop cluster? Regards Hans-Peter *From:* Yang Song [mailto:xfil...@gmail.com] *Sent:* donderdag 10 januari 2013 19:22 *To:* user@cassandra.apache.org *Subject:* Re

Re: Starting Cassandra

Could you also let us know if switching openjdk to jdk@oracle indeed solves the problem? Thanks! Yang 2013/1/10 Sloot, Hans-Peter hans-peter.sl...@atos.net I have increased the memory to 4096. Did not help It is openjdk indeed. java-1.6.0-openjdk.x86_64 1

LOCAL_QUORUM vs EACH_QUORUM

code, is there a way to achieve that? if not, creating a new policy does not seem too difficult either. Thanks Yang

Re: replace_token code?

replied in blue, Thanks Yang I thought the very first log line already acquired ownership , instead of later in the sequence? WARN [main] 2012-09-10 08:00:21,855 TokenMetadata.java (line 160) Token 166594924822352415786406422619018814804 changing ownership from / 10.72.201.80

replace_token code?

80 forcefully from gossip. instead, a few seconds later, it believed that .80 became live again. I don't have much understanding of the Gossip protocol, but roughly know that it's probability-based, looks we need an assertive/NOW membership control message for replace_token. thanks yang WARN

Re: replace_token code?

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

keyspace not propagated to a new node?

not see the KS. do I have to do something? I thought the bootstrap process would get the KS'es??? also I tried the -Dcassandra.replace_token, same issue thanks Yang

Re: keyspace not propagated to a new node?

this is 1.1.2 unfortunately the logs does not show anything except for the initial startup sequences. on the siblings, somehow the /var/log/cassandra/system.log were wrongly deleted, so I can't see their responses either I'll try resetlocalschema next time thanks! yang On Tue, Sep 4, 2012 at 7

Re: how is this possible ---- token range sum 100% ?

/CASSANDRA-3412 [2] https://github.com/apache/cassandra/blob/trunk/src/java/org/apache/cassandra/service/StorageServiceMBean.java#L342 On Wed, Aug 29, 2012 at 7:02 PM, Yang tedd...@gmail.com wrote: I have 16 nodes, each of them should have only 6.25%, but now they all show 18.73% how

Re: JMX(RMI) dynamic port allocation problem still exists?

nice, thanks guys! On Tue, Aug 28, 2012 at 11:58 PM, Senthilvel Rangaswamy senthil...@gmail.com wrote: We use Jolokia. Its pretty awesome. On Tue, Aug 28, 2012 at 11:44 PM, Peter Schuller peter.schul...@infidyne.com wrote: I can recommend Jolokia highly for providing an HTTP/JSON

a simple solution to replace dead node

launching of the new host. but that limitation is due to nventor/Puppet. Just want to share with the community here, cuz many people probably run Puppet+nventory, so this simple solution might work well Thanks Yang

Re: JMX(RMI) dynamic port allocation problem still exists?

/redir.aspx?C=PLVE6taKpU--Dxw69WVEOtdUcArCWM8IUH6LBjdXcM7STlqwkARq8mA8Nva_mtGaSKmmva4pWxE.URL=http%3a%2f%2fwww.datastax.com%2fdocs%2f1.1%2finstall%2finstall_ami Yang On Mon, Aug 27, 2012 at 9:03 PM, Nick Bailey n...@datastax.com wrote: The problem still exists. There was a discussion about

JMX(RMI) dynamic port allocation problem still exists?

? Thanks Yang

Re: JMX(RMI) dynamic port allocation problem still exists?

dean.hil...@nrel.gov wrote: In cassandra-env.sh, search on JMX_PORT and it is set to 7199 (ie. Fixed) so that solves your issue, correct? Dean From: Yang tedd...@gmail.commailto:tedd...@gmail.com Reply-To: user@cassandra.apache.orgmailto:user@cassandra.apache.org user

nodetool output through REST API?

REST API or python API, so that I avoid the possible changes in formatting of the output? I checked pycassa, it doesn't seem to have an API for the JMX services Thanks Yang

Re: replace dead node? token -1

Thanks Yang On Tue, Aug 14, 2012 at 10:17 PM, Jim Cistaro jcist...@netflix.com wrote: We use priam to replace nodes using replace_token. We do see some issues (currently on 1.0.9, as well as earlier versions) with replace_token. Apparently there are some known issues with replace_token. We have

Re: replace dead node? token -1

it's another issue? Thanks Yang On Tue, Aug 14, 2012 at 11:03 PM, Yang tedd...@gmail.com wrote: Jim: thanks a lot for the info. when you say old nodes sometimes hanging around as unreachable nodes when describing cluster, you mean after the new node boots up and assumes ownership

Re: replace dead node? token -1

of partitioning the data range. Thanks Yang On Wed, Aug 15, 2012 at 1:39 AM, Yang tedd...@gmail.com wrote: ok, I see, the cassandra.replace_token setting essentially executes the manual removeToken step. so the dead node should be removed. is this the old node hanging around issue that you

Re: replace dead node? token -1

be more attractive for you. We don't use it, so I cannot say if that approach has any issues, etc. Jim From: Yang tedd...@gmail.com Reply-To: user@cassandra.apache.org Date: Wed, 15 Aug 2012 02:00:55 -0700 To: user@cassandra.apache.org Subject: Re: replace dead node? token -1

incremental backup and Priam?

pretty much does the same thing as the original addSSTable() incremental backup . so the Priam backup code operates outside of Cassandra write path? any insight into why this approach was chosen instead of using the incremental backup provided by Cassandra? thanks Yang

Re: incremental backup and Priam?

taking the built in incremental backups off node. (AFAIK) Cheers - Aaron Morton Freelance Developer @aaronmorton http://www.thelastpickle.com On 15/08/2012, at 8:16 AM, Yang tedd...@gmail.com wrote: in the initial incremental backup implementation, the hardlinking

replace dead node? token -1

, but assigns the original token T to the new node. if so, would the duplicate token (same token but different ip) cause problems? Thanks Yang

Re: replace dead node? token -1

- Aaron Morton Freelance Developer @aaronmorton http://www.thelastpickle.com On 15/08/2012, at 11:07 AM, Yang tedd...@gmail.com wrote: previously when a node dies, I remember the documents describes that it's better to assign T-1 to the new node, where T was the token

Re: how to debug/trace

normally I'd just fire up debug in eclipse, make a break point on the Cassandra.server methods. On Fri, Dec 16, 2011 at 2:19 PM, S Ahmed sahmed1...@gmail.com wrote: How can you possibly trace a read/write in cassandra's codebase when it uses so many threadpools/executers? I'm just getting

Re: simple get_slice() gives error?

is no zero, possibly due to the slab allocator. so do you remember if there is some place in the Cassandra readpath that always copies the ByteBuffer to one that starts a 0 ? or I'm having some bug incidences I haven't noticed? Thanks Yang On Fri, Jun 10, 2011 at 7:09 PM, Jonathan Ellis jbel

Re: weird problem with performance

actually this is only an issue in HH, since HH writes all the stored messages into the same row, so locking is a problem 2011/10/21 Jérémy SEVELLEC jsevel...@gmail.com: @Araron you're right and i was wrong! 2011/10/20 Yang tedd...@gmail.com found it , https://issues.apache.org/jira

Re: SimpleAuthenticator / SimpleAuthorization missing

See: https://issues.apache.org/jira/browse/CASSANDRA-2922 On Thu, Oct 20, 2011 at 4:08 AM, Pierre Chalamet pie...@chalamet.netwrote: Hello, SimpleAuthenticator SimpleAuthorization just disappear in release 1.0.0... Will this stay like this or is it a release bug ? Thanks, - Pierre

Re: weird problem with performance

 to achieve QUORUM when you read in that context. It may be one explanation. You can turn cassandra log into debug level to see what happen when when there is a down node. 2011/10/19 Yang tedd...@gmail.com 3 sorry forgot this important info On Oct 19, 2011 11:31 AM, Jérémy SEVELLEC jsevel

weird problem with performance

for this?(or how to debug this?) Thanks Yang

Re: weird problem with performance

3 sorry forgot this important info On Oct 19, 2011 11:31 AM, Jérémy SEVELLEC jsevel...@gmail.com wrote: Hi, what is your replication_factor? 2011/10/19 Yang tedd...@gmail.com I'm using a cassandra version compiled from 1.0.0 github HEAD. I have 3 nodes, A B and C, on node A I run

DELETE where colname == given_value ?

there is not a way to do this in batch. is it possible to add this logic to a flow like get_range_slice, which scans through the sstables ? Thanks Yang

commitlog replay extremely slow?

on 4f39d3e52f82d060bf96c2be0df6ff6782bc48e5 ? those changes after this do not sound immediately related to the possible issue I'm seeing Thanks Yang

Re: commitlog replay extremely slow?

; if (mutatedIndexedColumns != null) it's a bit weird, since I checked that indexLockFor() does return distinct objects from the array of 4096 objects. but every time I do a jstack, it shows them in this state, what are they blocking on ?? On Tue, Oct 18, 2011 at 4:37 PM, Yang tedd...@gmail.com wrote

add bloomfilter results to nodetool?

I find the info about bloomfilter very helpful, could we add that to NodeCmd ? Thanks Yang

different size sstable on different nodes?

to figure out, just want to see if there are any apparent things I missed thanks Yang

Re: different size sstable on different nodes?

, but I just finished running the compaction, and GC_SECONDS is 7200 , set short for testing purpose. so this deletion column should have been thrown away during the compaction On Tue, Oct 11, 2011 at 10:33 AM, Yang tedd...@gmail.com wrote: after I did a major compaction on both nodes in my

Re: ebs or ephemeral

Agree, EBS systems are not so good for cassandra systems and during previous conversations in this mail list, people tend to use ephemeral. 從我的 BlackBerry® 無線裝置 -Original Message- From: Sasha Dolgy sdo...@gmail.com Date: Mon, 10 Oct 2011 10:03:26 To: user@cassandra.apache.org Reply-To:

factors on the effectiveness of bloom filter?

not be changed (or the only possibility to change they key is simply juggle the byte order), is there any measure to increase the effectiveness of bloom filters? thanks Yang

Re: ebs or ephemeral

Obviously ephemeral. It has higher IO availability, will not affect your Ethernet IO performance, and it is free (included in instance price) and the redundancy is provided by cassandra itself. 從我的 BlackBerry® 無線裝置 -Original Message- From: Madalina Matei madalinaima...@gmail.com Date:

Re: Why is mutation stage increasing ??

Well what client are you using? And can you give a hint to your node hardware? 從我的 BlackBerry® 無線裝置 -Original Message- From: Philippe watche...@gmail.com Date: Wed, 5 Oct 2011 10:33:21 To: useruser@cassandra.apache.org Reply-To: user@cassandra.apache.org Subject: Why is mutation stage

EC2 raid0 disks ?

it seems that how many virtual disks you can have is fixed: on m2.4xlarge you have 2 disks, while on m2.2xlarge you have only 1, so I can't setup a raid0 on m2.2xlarge am I correct? Thanks Yang

Re: EC2 raid0 disks ?

are all virtual, would there be any benefit at all in doing a raid0 ? Yang 2011/10/4 Joaquin Casares joaq...@datastax.com: Hello again, Also, EBS volumes can be attached, but the performance issues cause other issues when running a healthy cluster. From experience running clusters on EBS

Re: cassandra performance degrades after 12 hours

maybe try row cache ? have you enabled the mlock ? (need jna.jar , and set ulimit -l ) using iostat -x would also give you more clues as to disk performance On Mon, Oct 3, 2011 at 10:12 AM, Ramesh Natarajan rames...@gmail.com wrote: I am running a cassandra cluster of  6 nodes running RHEL6

disable mysterious GC

the following source code in jdk , RMI part, forces a full gc every 1 hour , if no old gen gc has happened by then. /** maximum interval between complete garbage collections of local heap */ private static final long gcInterval = // default 1 hour

Re: disable mysterious GC

and server=Cassandra.  But sun's docs say that sun.rmi.dgc.server.gcInterval defaults to 60s which I am definitely NOT seeing. On Mon, Oct 3, 2011 at 4:12 PM, Yang tedd...@gmail.com wrote: the following source code in jdk , RMI part, forces a full gc every 1 hour , if no old gen gc has happened

Re: disable mysterious GC

btw the first code snippet is from openjdk 7 On Mon, Oct 3, 2011 at 2:29 PM, Yang tedd...@gmail.com wrote: looks doc is outdated : $ grep '\.gcInterval' ./j2se/src/share/classes/sun/rmi/transport/ObjectTable.java            new GetLongAction(sun.rmi.dgc.server.gcInterval, 360

Re: Cassandra JVM heap size

Someone has just talked about the heap size in this mail list, says that bigger heap size will result into a longer GC phase, that could probably be one of the reason not using larger heap size. But I have really heard of some others using Cassandra with some 60 gigabytes of heap size. 從我的

release mmap memory through jconsole?

I gave an -Xmx50G to my Cassandra java processs, now top shows its virtual memory address space is 82G, is there a way to release that memory through JMX ? Thanks Yang

Re: release mmap memory through jconsole?

It is meaningless to release such memory. The counting includes the data you reached in the SSTable. Those data locates on your hard drive. So it is not the RAM spaces you have actually used. -Y. --Original Message-- From: Yang To: user@cassandra.apache.org ReplyTo: user

Re: release mmap memory through jconsole?

people will tell you that 12G -16G is max to use. Bye, Norman 2011/9/30 Yi Yang i...@iyyang.com: It is meaningless to release such memory. The counting includes the data you reached in the SSTable. Those data locates on your hard drive. So it is not the RAM spaces you have actually used. -Y

Re: release mmap memory through jconsole?

access pattern is random though, otherwise OS cache has the benefit of spacial locality (it loads blocks at once instead of a single row). thanks Yang 2011/9/30 Norman Maurer norman.mau...@googlemail.com: I would also not use such a big heap. I think most people will tell you that 12G -16G

Re: release mmap memory through jconsole?

you're seeing lots of young gen-based GC pauses... :) 2011/9/30 Yang tedd...@gmail.com: why? I thought bigger young gen would allow more objects to die (become non-reachable) before minor collection, so the minor collection cost is low. particularly it would allow you to merge more updates

  1   2   3   >