Re: unified highlighter performance in solr 8.5.1

2020-07-04 Thread David Smiley
Here's my PR, which includes some edits to the ref guide docs where I tried to clarify these settings a little too. https://github.com/apache/lucene-solr/pull/1651 ~ David On Sat, Jul 4, 2020 at 8:44 AM Nándor Mátravölgyi wrote: > I guess that's fair. Let's have hl.fragsizeIsMinimum=true as def

Re: Time-out errors while indexing (Solr 7.7.1)

2020-07-04 Thread Mad have
Thank a lot for your inputs and suggestions, even I was thinking similar like creating another collection of the same ( hot and cold), and moving documents which are older than certain days like 180 days from original collection (hot) to new collection(cold). Thanks, Madhava Sent from my iPh

Re: Almost nodes in Solrcloud dead suddently

2020-07-04 Thread Tran Van Hoan
*total thread is 25.6k when solr hang. On Sunday, July 5, 2020, 2:55:26 AM GMT+7, Tran Van Hoan wrote: All server only run Solr, zookeeper, exporters (node-exporter, process-exporter, solr-exporter, zoo-exporter). - network: no package loss, TCP no issue before incident, TCP drop ar

Re: Almost nodes in Solrcloud dead suddently

2020-07-04 Thread Tran Van Hoan
All server only run Solr, zookeeper, exporters (node-exporter, process-exporter, solr-exporter, zoo-exporter). - network: no package loss, TCP no issue before incident, TCP drop around 100-200/s when incident and overflow ~100 in somaxcon. - total mem Avalilable is greater 25G (Solr's XMX = 30G

Re: Almost nodes in Solrcloud dead suddently

2020-07-04 Thread Rodrigo Oliveira
Network it's ok? Between nodes? The use? Swap it's disabled? Swapiness rhe value it's 0? Em sáb, 4 de jul de 2020 15:19, Tran Van Hoan escreveu: > I used physical servers, and IO wait is small :(!!!I saw that iptables > dropped all ACK message from clients (not only client solr, prometheus > sc

Re: Almost nodes in Solrcloud dead suddently

2020-07-04 Thread Tran Van Hoan
I used physical servers, and IO wait is small :(!!!I saw that iptables dropped all ACK message from clients (not only client solr, prometheus scape metric from exporter was dropped too).all when i check netstat -anp|grep 8983, all socket are TIME_WAIT state.Only restart solrs, the incident was

Re: Almost nodes in Solrcloud dead suddently

2020-07-04 Thread Rodrigo Oliveira
Hi, I had this problem. In my case was the wait/io in vm. I migrate my environment to another place and solved. Actually it's problem wirh wait/io at host physical (until backup it's a problem over veeam). Regards Em sáb, 4 de jul de 2020 12:30, Tran Van Hoan escreveu: > The problem reoccurs

Re: Almost nodes in Solrcloud dead suddently

2020-07-04 Thread Tran Van Hoan
The problem reoccurs repeatly in recent days. To day i tried dump heap and thread. Only dumping thread, heap can not because solr instance was hang.Almost thread was blocked. On Tuesday, June 23, 2020, 10:42:36 PM GMT+7, Tran Van Hoan wrote: I checked node exporter metrics and saw n

Re: Questions about Solr Search

2020-07-04 Thread Nicolas Franck
Short answer: no Neither Solr nor ElasticSearch have such capabilities out of the box. Solr does have a plugin infrastructure that enables you to provide better tokenization based on language rules, and some are better than others. I saw for example integration of openNLP here: https://lucene.a

Re: Time-out errors while indexing (Solr 7.7.1)

2020-07-04 Thread Erick Erickson
You need more shards. And, I’m pretty certain, more hardware. You say you have 13 billion documents and 6 shards. Solr/Lucene has an absolute upper limit of 2B (2^31) docs per shard. I don’t quite know how you’re running at all unless that 13B is a round number. If you keep adding documents, you

Re: unified highlighter performance in solr 8.5.1

2020-07-04 Thread Nándor Mátravölgyi
I guess that's fair. Let's have hl.fragsizeIsMinimum=true as default. On 7/4/20, David Smiley wrote: > I doubt that WORD mode is impacted much by hl.fragsizeIsMinimum in terms of > quality of the highlight since there are vastly more breaks to pick from. > I think that setting is more useful in S

Re: Time-out errors while indexing (Solr 7.7.1)

2020-07-04 Thread Mad have
Hi Eric, There are total 6 VM’s in Solr clusters and 2 nodes are running on each VM. Total number of shards are 6 with 3 replicas. I can see the index size is more than 220GB on each node for the collection where we are facing the performance issue. The more documents we add to the collection

***URGENT***Re: Questions about Solr Search

2020-07-04 Thread Gautam K
Since it's a bit of an urgent request so if could please help me on this by today it will be highly appreciated. Thanks & Regards, Gautam Kanaujia On Thu, Jul 2, 2020 at 7:49 PM Gautam K wrote: > Dear Team, > > Hope you all are doing well. > > Can you please help with the following question? We

Questions about Solr Search

2020-07-04 Thread Gautam K
Dear Team, Hope you all are doing well. Can you please help with the following question? We are using Solr search in our Organisation and now checking whether Solr provides search capabilities like Google Enterprise search(Google Knowledge Graph Search). 1, Does Solr Search provide Voice Search