I thought something similar but wasn't sure - but this answers it. On Wednesday, January 7, 2015 1:09:50 PM UTC-8, Mark Walkom wrote: > > Generally - Data needs RAM, disk and CPU, client would need CPU and RAM, > master can be super light (4-8GB, a single core). > > On 8 January 2015 at 07:10, Bhumir Jhaveri <bhum...@gmail.com > <javascript:>> wrote: > >> Oh nice - so some indexes which I have created month ago which has - lets >> say 5 shards on one node - after adding node - 2 of these existing shards >> might move to newly added data node - this is great! >> >> So when you create master, client, data node - out of these which one >> needs to be superly powerful in terms of RAM/CPU/SPACE - I agree that data >> nodes need to have superior space since they will actually store the index >> data but not sure which one should be powerful interms of RAM/CPU or all of >> them should be equally good like 12 CPU, 64 gigs of ram etc? >> >> I agree it purely depends on whats the volume of data etc but in general >> who should be most powerful in terms of hardware capacity? >> >> >> >> On Wednesday, January 7, 2015 11:46:57 AM UTC-8, Mark Walkom wrote: >>> >>> Yes it auto distributes existing, and new, shards. >>> >>> On 8 January 2015 at 05:55, Bhumir Jhaveri <bhum...@gmail.com> wrote: >>> >>>> Also one more question - lets say intially I have one node architecture >>>> - i.e. everything on one single node and additional mount is having all >>>> the >>>> ES data (index, documents etc) which is like of 300 gigs - now lets say if >>>> I add more data nodes - so going forward it will distribute the shards to >>>> different nodes or existing data will automatically be redistributed too? >>>> >>>> >>>> >>>> On Wednesday, January 7, 2015 9:57:44 AM UTC-8, David Pilato wrote: >>>>> >>>>> No you don’t have to reindex. >>>>> Elasticsearch can read segments generated with previous elasticsearch >>>>> version. >>>>> >>>>> -- >>>>> *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 7 janv. 2015 à 18:50, Bhumir Jhaveri <bhum...@gmail.com> a écrit : >>>>> >>>>> Alright cool David! if this is the pain point then I dont think that I >>>>> should limit this upgrade to only 1.3 - I will go for 1.4 >>>>> >>>>> one more thing - I already have around 300 GB(max capacity 2TB) of >>>>> indexed data available in my additional storage which I exclusively kept >>>>> for ES - so far this 300 gigs of data has been generated by ES1.2 - now >>>>> if >>>>> I move to ES 1.4 - will that cause any issue? >>>>> Do I need to regenerate anything over here or ES 1.4 will >>>>> automatically accept whatever has been generated by ES1.2? >>>>> >>>>> its more of back compatibility point which I am trying to raise here. >>>>> >>>>> >>>>> >>>>> On Wednesday, January 7, 2015 9:46:50 AM UTC-8, David Pilato wrote: >>>>>> >>>>>> It’s most likely because on your dev system, you are running out of >>>>>> disk space. >>>>>> Elasticsearch 1.4.x does not allocate replicas if you have more than >>>>>> 85% disk usage. >>>>>> You can change this settings by modifying for example >>>>>> elasticsearch.yml and set: >>>>>> >>>>>> cluster.routing.allocation.disk.watermark.low: 1gb >>>>>> cluster.routing.allocation.disk.watermark.high: 500mb >>>>>> >>>>>> Regarding your question, I would go for 1.4. >>>>>> >>>>>> HTH >>>>>> >>>>>> -- >>>>>> *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 7 janv. 2015 à 18:41, Bhumir Jhaveri <bhum...@gmail.com> a écrit : >>>>>> >>>>>> I am planning to upgrade from ES 1.2.2 to ES 1.4.2 or 1.3.9 or >>>>>> whichever is the most latest (stable) in 1.3 series - so should I go for >>>>>> 1.4 or should stay with 1.3? >>>>>> the reason why I am not thinking about absolute upgrade to 1.4 is >>>>>> just so that if certain things like shards allocations to different >>>>>> node/s >>>>>> suddenly doesnt work or may be I am wrong too - I mean this type of >>>>>> sudden >>>>>> and obvious things just stops working - I am still new to ES world and >>>>>> not >>>>>> knowing ES fully. >>>>>> >>>>>> Not sure - need opinions on your experience if any? >>>>>> >>>>>> >>>>>> -- >>>>>> 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 elasticsearc...@googlegroups.com. >>>>>> To view this discussion on the web visit https://groups.google.co >>>>>> m/d/msgid/elasticsearch/ae8afb06-d923-4bf6-ac3c-adcb35a51fdf >>>>>> %40googlegroups.com >>>>>> <https://groups.google.com/d/msgid/elasticsearch/ae8afb06-d923-4bf6-ac3c-adcb35a51fdf%40googlegroups.com?utm_medium=email&utm_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 elasticsearc...@googlegroups.com. >>>>> To view this discussion on the web visit https://groups.google.co >>>>> m/d/msgid/elasticsearch/5e503a64-c9b2-42f2-9d0f-187c89f097dc >>>>> %40googlegroups.com >>>>> <https://groups.google.com/d/msgid/elasticsearch/5e503a64-c9b2-42f2-9d0f-187c89f097dc%40googlegroups.com?utm_medium=email&utm_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 elasticsearc...@googlegroups.com. >>>> To view this discussion on the web visit https://groups.google.com/d/ >>>> msgid/elasticsearch/b2818693-9307-44fc-af38-b529d2fc3d05% >>>> 40googlegroups.com >>>> <https://groups.google.com/d/msgid/elasticsearch/b2818693-9307-44fc-af38-b529d2fc3d05%40googlegroups.com?utm_medium=email&utm_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 elasticsearc...@googlegroups.com <javascript:>. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/elasticsearch/9b6ea2b8-4f18-4516-9178-d9b9fc428e92%40googlegroups.com >> >> <https://groups.google.com/d/msgid/elasticsearch/9b6ea2b8-4f18-4516-9178-d9b9fc428e92%40googlegroups.com?utm_medium=email&utm_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/187fd43e-8459-4bfd-be73-1300afc60d23%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
