Re: Shard Balancing
You are right. Elasticsearch distribute shards to 10 nodes most of the time. But there are exceptions and I want to guarantee it that every time shards will be distributed even or at least decrease frequency of these exceptions. On 14-01-2015 13:10, David Pilato wrote: If you start 10 nodes and have 10 shards to allocate, elasticsearch will most likely do what you are expecting: 1 shard per node. So you don’t really need to define any setting but just use elasticsearch defaults which are good most of the time. -- *David Pilato* | /Technical Advocate/ | *Elasticsearch.com http://Elasticsearch.com* @dadoonet https://twitter.com/dadoonet | @elasticsearchfr https://twitter.com/elasticsearchfr |@scrutmydocs https://twitter.com/scrutmydocs Le 14 janv. 2015 à 10:59, Umutcan umut...@gamegos.com mailto:umut...@gamegos.com a écrit : Hi, I have a question regarding balance settings of Elasticsearch. I have 10 nodes and each index has 10 shards (5 primary, 5 replica). I want for all nodes to have only 1 shard of an index. The documents explains some settings here x-msg://4/elasticsearch@googlegroups.com to do this, but I am not sure what values to use. Is it a good idea to set 1.0 first three settings? Is there any possible problems if I change these settings? Best, Umutcan -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com mailto:elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/54B63E16.5050608%40gamegos.com https://groups.google.com/d/msgid/elasticsearch/54B63E16.5050608%40gamegos.com?utm_medium=emailutm_source=footer. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com mailto:elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/05274FAC-8C3B-41DA-A0B5-6F47A6589C45%40pilato.fr https://groups.google.com/d/msgid/elasticsearch/05274FAC-8C3B-41DA-A0B5-6F47A6589C45%40pilato.fr?utm_medium=emailutm_source=footer. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/54B660BB.9020202%40gamegos.com. For more options, visit https://groups.google.com/d/optout.
Unassigned Shards Of A Newer Elasticsearh Version
Hi, When I was doing some experiments on my Elasticsearch cluster, I found that cluster cannot assign replicas of a shards on the node which have a new version of Elasticsearch (1.4.1). I have seen this before and I can fix this by upgrading old nodes. But, I am wondering that if there is a way to fix this issue without upgrade. Is there any idea or a solution? Thanks, Umutcan PS: Other nodes in the cluster have Elasticsearch 1.4. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/547C251B.9060809%40gamegos.com. For more options, visit https://groups.google.com/d/optout.
Re: Changed Master Node and New Master Did Not Find Old Indices
I am using 1.4.0 version. There is nothing in the logs. I started new master after I shut down the old one. I thought new master could detect indices, but I guess master node stores some meta data related to indices and these data is gone with the old master node. New master created all the indices from scratch. I use timestamp in the index names, so indices before yesterday is still remains in the data nodes. I want to load these indices. On 19-11-2014 12:36, Mark Walkom wrote: What version are you on? Is there anything in the logs? On 19 November 2014 18:14, Umutcan umut...@gamegos.com mailto:umut...@gamegos.com wrote: Yes, it did and both are set to true. But, it did not join the cluster until I shut down the old master. On 18-11-2014 22:17, Mark Walkom wrote: Did it join the cluster without any problems? Is it set to master: true and data:true? On 19 November 2014 02:29, Umutcan umut...@gamegos.com mailto:umut...@gamegos.com wrote: Hi, I have a cluster with 1 master and 3 data nodes. I needed to change master node. So, I created a new master node and shutdown the old one. Both are EC2 instances. After doing this new master did not find old indices. All the files belong to old indices still remains inside the data nodes. Is there a way to make new master node find those files and use old indices? Thanks, Umutcan -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com mailto:elasticsearch%2bunsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/546B65DB.1040201%40gamegos.com. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com mailto:elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/CAF3ZnZmspn9Nh0V7YZGkhYXFBKWR1B%3DYonZm1jpk%2By5oivcKNw%40mail.gmail.com https://groups.google.com/d/msgid/elasticsearch/CAF3ZnZmspn9Nh0V7YZGkhYXFBKWR1B%3DYonZm1jpk%2By5oivcKNw%40mail.gmail.com?utm_medium=emailutm_source=footer. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com mailto:elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/546C435F.102%40gamegos.com https://groups.google.com/d/msgid/elasticsearch/546C435F.102%40gamegos.com?utm_medium=emailutm_source=footer. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com mailto:elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/CAF3ZnZnj9Hv7gvwOXJNifOYJsksqFb65n%3Dt5uMkYwzmgxk3ZwQ%40mail.gmail.com https://groups.google.com/d/msgid/elasticsearch/CAF3ZnZnj9Hv7gvwOXJNifOYJsksqFb65n%3Dt5uMkYwzmgxk3ZwQ%40mail.gmail.com?utm_medium=emailutm_source=footer. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/546C8F55.5000700%40gamegos.com. For more options, visit https://groups.google.com/d/optout.
Re: Changed Master Node and New Master Did Not Find Old Indices
You are right. It should be ok if the new node joined before I remove old one, but the new master joined cluster after I removed the old one. I will change my cluster configuration. I think it is better to make all nodes both data and master. But, my question still remains. Is there a way to recover those indices from files? On 20-11-2014 00:42, Mark Walkom wrote: If your new master capable node joined the cluster then it would have received the cluster state metadata, and (presuming things are all ok) when you remove the old master then it should be promoted and take over without problems. I don't really know much about the AWS plugin though, but a standard cluster works in that manner. On 19 November 2014 23:38, Umutcan umut...@gamegos.com mailto:umut...@gamegos.com wrote: I am using 1.4.0 version. There is nothing in the logs. I started new master after I shut down the old one. I thought new master could detect indices, but I guess master node stores some meta data related to indices and these data is gone with the old master node. New master created all the indices from scratch. I use timestamp in the index names, so indices before yesterday is still remains in the data nodes. I want to load these indices. On 19-11-2014 12:36, Mark Walkom wrote: What version are you on? Is there anything in the logs? On 19 November 2014 18:14, Umutcan umut...@gamegos.com mailto:umut...@gamegos.com wrote: Yes, it did and both are set to true. But, it did not join the cluster until I shut down the old master. On 18-11-2014 22:17, Mark Walkom wrote: Did it join the cluster without any problems? Is it set to master: true and data:true? On 19 November 2014 02:29, Umutcan umut...@gamegos.com mailto:umut...@gamegos.com wrote: Hi, I have a cluster with 1 master and 3 data nodes. I needed to change master node. So, I created a new master node and shutdown the old one. Both are EC2 instances. After doing this new master did not find old indices. All the files belong to old indices still remains inside the data nodes. Is there a way to make new master node find those files and use old indices? Thanks, Umutcan -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com mailto:elasticsearch%2bunsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/546B65DB.1040201%40gamegos.com. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com mailto:elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/CAF3ZnZmspn9Nh0V7YZGkhYXFBKWR1B%3DYonZm1jpk%2By5oivcKNw%40mail.gmail.com https://groups.google.com/d/msgid/elasticsearch/CAF3ZnZmspn9Nh0V7YZGkhYXFBKWR1B%3DYonZm1jpk%2By5oivcKNw%40mail.gmail.com?utm_medium=emailutm_source=footer. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com mailto:elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/546C435F.102%40gamegos.com https://groups.google.com/d/msgid/elasticsearch/546C435F.102%40gamegos.com?utm_medium=emailutm_source=footer. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com mailto:elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/CAF3ZnZnj9Hv7gvwOXJNifOYJsksqFb65n%3Dt5uMkYwzmgxk3ZwQ%40mail.gmail.com https://groups.google.com/d/msgid/elasticsearch/CAF3ZnZnj9Hv7gvwOXJNifOYJsksqFb65n%3Dt5uMkYwzmgxk3ZwQ%40mail.gmail.com
Changed Master Node and New Master Did Not Find Old Indices
Hi, I have a cluster with 1 master and 3 data nodes. I needed to change master node. So, I created a new master node and shutdown the old one. Both are EC2 instances. After doing this new master did not find old indices. All the files belong to old indices still remains inside the data nodes. Is there a way to make new master node find those files and use old indices? Thanks, Umutcan -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/546B65DB.1040201%40gamegos.com. For more options, visit https://groups.google.com/d/optout.
Re: Changed Master Node and New Master Did Not Find Old Indices
Yes, it did and both are set to true. But, it did not join the cluster until I shut down the old master. On 18-11-2014 22:17, Mark Walkom wrote: Did it join the cluster without any problems? Is it set to master: true and data:true? On 19 November 2014 02:29, Umutcan umut...@gamegos.com mailto:umut...@gamegos.com wrote: Hi, I have a cluster with 1 master and 3 data nodes. I needed to change master node. So, I created a new master node and shutdown the old one. Both are EC2 instances. After doing this new master did not find old indices. All the files belong to old indices still remains inside the data nodes. Is there a way to make new master node find those files and use old indices? Thanks, Umutcan -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com mailto:elasticsearch%2bunsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/546B65DB.1040201%40gamegos.com. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com mailto:elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/CAF3ZnZmspn9Nh0V7YZGkhYXFBKWR1B%3DYonZm1jpk%2By5oivcKNw%40mail.gmail.com https://groups.google.com/d/msgid/elasticsearch/CAF3ZnZmspn9Nh0V7YZGkhYXFBKWR1B%3DYonZm1jpk%2By5oivcKNw%40mail.gmail.com?utm_medium=emailutm_source=footer. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/546C435F.102%40gamegos.com. For more options, visit https://groups.google.com/d/optout.
Unassigned Shards and Upgrading Elasticsearch
Hi, Yesterday, I have added a new node to my ES cluster and after that some of the shards started to remain unassigned. These shards are the replicas of the shards on the new node. When I inspected the reason, I found out that this new node has a different ES version. The new one is 1.3.4 and the older ones are 1.3.2. Does this difference cause the problem? If I upgrade the older ones, does this solves the issue? I installed ES via apt-get in Ubuntu. If I upgrade the current nodes via apt-get, does this cause problems with old indices? Thanks, Umutcan -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/545B3152.1080606%40gamegos.com. For more options, visit https://groups.google.com/d/optout.
A Problem About UDP port 80
Hi, We have been testing Elasticsearch for a while. Our ES cluster was on AWS. We installed Bigdesk, Marvel, Thrift, EC2 Discovery plugins. There were 5 instance (1 load balancer, 4 data node) and all of them were version 0.90. Yesterday, We have received an e-mail from AWS. They said one of our instance in ES cluster was making DOS attacks from UDP port 80. We did not restrict ports, because it was an test cluster. It can be main cause of this problem, but I still want to ask if there is a known bug (in ES or modules or plugins) that cause something like this or if there is anyone who have seen some kind of similar problem. Thanks, Umutcan Onal -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/53BD012A.2090109%40gamegos.com. For more options, visit https://groups.google.com/d/optout.
Re: A Problem About UDP port 80
You are right. We were aware that exposing cluster to internet was a bad idea. It was a temporary situation. We are planing to use it behind an application in our product. Thanks for advice. On 09-07-2014 12:01, David Pilato wrote: 3 bad things here: * You exposed your cluster to internet directly * You did not disable dynamic scripting * May be you are running your elasticsearch node as root? You should read that documentation: http://www.elasticsearch.org/guide/en/elasticsearch/reference/0.90/modules-scripting.html#_disabling_dynamic_scripts -- *David Pilato* | /Technical Advocate/ | *Elasticsearch.com* @dadoonet https://twitter.com/dadoonet | @elasticsearchfr https://twitter.com/elasticsearchfr Le 9 juillet 2014 à 10:45:30, Umutcan (umut...@gamegos.com mailto:umut...@gamegos.com) a écrit: Hi, We have been testing Elasticsearch for a while. Our ES cluster was on AWS. We installed Bigdesk, Marvel, Thrift, EC2 Discovery plugins. There were 5 instance (1 load balancer, 4 data node) and all of them were version 0.90. Yesterday, We have received an e-mail from AWS. They said one of our instance in ES cluster was making DOS attacks from UDP port 80. We did not restrict ports, because it was an test cluster. It can be main cause of this problem, but I still want to ask if there is a known bug (in ES or modules or plugins) that cause something like this or if there is anyone who have seen some kind of similar problem. Thanks, Umutcan Onal -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/53BD012A.2090109%40gamegos.com. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com mailto:elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/etPan.53bd04d9.189a769b.6455%40MacBook-Air-de-David.local https://groups.google.com/d/msgid/elasticsearch/etPan.53bd04d9.189a769b.6455%40MacBook-Air-de-David.local?utm_medium=emailutm_source=footer. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/53BD0DD2.5080400%40gamegos.com. For more options, visit https://groups.google.com/d/optout.
JVM Memory Usage
Hi, I have a question about memory usage. My cluster has 1 master and data node and 3 data nodes. Each have 6 GB heap size (which is nearly half of the machine). I have 250 shards with replicas and I have 600 GB data in total. When I start the cluster, I can use it for 1 week without any problem. After a week, my cluster begins to fail due to low memory (below 10%). When I restart all the nodes, everything is fine, again. Free memory goes up to 40%. And, it fails again 1 week after the restart. I think some data is remaining in the memory for a long time even if it is not used. Is there any configuration to optimize this? Do I need to flush indices or clear cache periodically? Thanks, Umutcan -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/5322B754.80104%40gamegos.com. For more options, visit https://groups.google.com/d/optout.
Re: Relation Between Heap Size and Total Data Size
So, I am wondering that is there any relationship between heap size and total data size? Is there any formula to determine heap size based on data size? You might want to check that you're not running out of file handles: http://www.elasticsearch.org/tutorials/too-many-open-files/ Thanks Dan. This article solves my problem. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/530EDFEF.1060505%40gamegos.com. For more options, visit https://groups.google.com/groups/opt_out.
Relation Between Heap Size and Total Data Size
Hi, I created a Elasticsearch cluster with 4 instance. Elasticsearch 0.90.10 is running all of them. Heap size is 6 GB for all the instances, so total heap size is 24 GB. I have 5 shard for each index and each shard has 1 replica. A new index is created for every day, so all indices have nearly same size. When total data size reaches around 100 GB (replicas are included), my cluster begins to fail to allocate some of the shards (status yellow). After I delete some old indices and restart all the nodes, everything is fine (status is green). If I do not delete some data, status eventually turns red. So, I am wondering that is there any relationship between heap size and total data size? Is there any formula to determine heap size based on data size? Thanks, Umutcan -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/530CB5FE.80203%40gamegos.com. For more options, visit https://groups.google.com/groups/opt_out.
Re: Relation Between Heap Size and Total Data Size
There is enough space on every machine. I looked in the logs and find out that org.apache.lucene.store.LockObtainFailedException: Lock obtain timed out: NativeFSLock@/ebs/elasticsearch/elasticsearch-0.90.10/data/elasticsearch/nodes/0/indices/logstash-2014.02.26/0/index/write.lock is what causes the shard fails to start. On 02/25/2014 05:29 PM, Randy wrote: Probably low on disc on at least one machine. Monitor disc usage. Also look in the logs and find out what error you are getting. Report back. Sent from my iPhone On Feb 25, 2014, at 7:25 AM, Umutcan umut...@gamegos.com wrote: Hi, I created a Elasticsearch cluster with 4 instance. Elasticsearch 0.90.10 is running all of them. Heap size is 6 GB for all the instances, so total heap size is 24 GB. I have 5 shard for each index and each shard has 1 replica. A new index is created for every day, so all indices have nearly same size. When total data size reaches around 100 GB (replicas are included), my cluster begins to fail to allocate some of the shards (status yellow). After I delete some old indices and restart all the nodes, everything is fine (status is green). If I do not delete some data, status eventually turns red. So, I am wondering that is there any relationship between heap size and total data size? Is there any formula to determine heap size based on data size? Thanks, Umutcan -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/530CB5FE.80203%40gamegos.com. For more options, visit https://groups.google.com/groups/opt_out. -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/530D9286.60300%40gamegos.com. For more options, visit https://groups.google.com/groups/opt_out.
Connecting ES Cluster using Thrift
Hi, I am a newbie about Elasticsearch. I am trying ES for some of my projects. I want to connect ES cluster using Thrift with python client. I installed plugin to all machines and restarted them. Then, I try to connect via Thrift and it gives me socket.timeout: timed out error. I can connect to my cluster without any problem using HTTP. Is there any additional configurations on server side or client side? Thanks, Umutcan -- You received this message because you are subscribed to the Google Groups elasticsearch group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/b0c110df-1580-46fb-9fa0-2cb656fb2036%40googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
