Re: Replication Index fetch failed

On Thu, Sep 1, 2016 at 6:05 PM, Arkadi Colson wrote: > ERROR - 2016-09-01 14:30:43.653; [c:intradesk s:shard1 r:core_node5 > x:intradesk_shard1_replica1] org.apache.solr.common.SolrException; Index > fetch failed :org.apache.solr.common.SolrException: Unable to download >

Re: ShardDoc.sortFieldValues are not exposed in v5.2.1

This was made public in https://issues.apache.org/jira/browse/SOLR-7968 which is released already in 5.5 On Fri, Sep 2, 2016 at 12:01 AM, tedsolr wrote: > I'm attempting to perform my own merge of IDs with a MergeStrategy in > v5.2.1. > I'm a bit hamstrung because the

Re: what to try next for replica that will not stay up.

My guess is that recovery is indeed successful but the leader is repeatedly marking this replicas as 'down' using what we call Leader-Initiated-Recovery or LIR. We need to understand why that is happening. Are there GC issues on this new node? Can we see the logs on the leader from a bit before

Feedback on Match Query Parser (for fixing multiterm synonyms and other things)

I wanted to solicit feedback on my query parser, the match query parser ( https://github.com/o19s/match-query-parser). It's a work in progress, so any thoughts from the community would be welcome. The point of this query parser is that it's not a query parser! Instead, it's a way of selecting

Always add the marker when elevating documents

Hi, I followed the instructions on the wiki (https://wiki.apache.org/solr/QueryElevationComponent) to add a QueryElevationComponent searchComponent in my Solr 4.10.2 server and it is working as expected. I saw in the documentation that it is possible to see which documents were elevated by

RE: what to try next for replica that will not stay up.

Well, thanks for your suggestion, it may well come to that. My impression the problem is with the shard in some way in that we have created the replica on other nodes and it has done the same thing. We went from 2 nodes to 4 to add processing capacity to the cluster a couple of weeks back, but

ShardDoc.sortFieldValues are not exposed in v5.2.1

I'm attempting to perform my own merge of IDs with a MergeStrategy in v5.2.1. I'm a bit hamstrung because the ShardFieldSortedHitQueue is not public. When trying to build my own priority queue I found out that the field sortFieldValues in ShardDoc is package restricted. Now, in v6.1 I see that

RE: Modifying fl in QParser

All, Thank you all for your responses. Our custom QParser identifies several of many dynamic fields for construction of the actual Lucene query. For instance, given a custom Solr request consisting of "q={!xyzQP}scid:500247", a new query is to be constructed using information from a SQL

Re: what to try next for replica that will not stay up.

This may be too simplistic of course, but what if you totally wipe Solr off that machine, re-install it from scratch, bring it up and let it join the cluster, then add the replica? If it was me (since I have the luxury of VM's) I'd turn it off and build a new VM from the ground up, but I get that

what to try next for replica that will not stay up.

Hi If anyone has any suggestions of things I could try to resolve my issue where one replica on one of my solcloud 6.0.1 shards refuses to stay up, I'd love to hear them. In fact, I'll get you something off your amazon wishlist, within reason, if you can solve this puzzle. Today we pruned

Re: Default Field Cache

Yes. Thanks. On 9/1/16 4:53 AM, Alessandro Benedetti wrote: Are you looking for this ? org/apache/solr/core/SolrConfig.java:243 CacheConfig conf = CacheConfig.getConfig(this, "query/fieldValueCache"); if (conf == null) { Map args = new HashMap<>(); args.put(NAME,

Re: How using fl in query affects query time

thanks alex. On Thu, Sep 1, 2016 at 6:54 PM, Alexandre Rafalovitch wrote: > I believe enableLazyFieldLoading setting is supposed to help with the > partial-fields use-case. But not with query time itself, but with > re-hydrating stored fields to return. Which I guess is part

Re: How using fl in query affects query time

I believe enableLazyFieldLoading setting is supposed to help with the partial-fields use-case. But not with query time itself, but with re-hydrating stored fields to return. Which I guess is part of the query time from the user's point of view.

How using fl in query affects query time

Hi, I am having around 100 fields in single document. I want to know that if I use fl and get only single field from query will that reduce query time?? or getting all the fields through query and getting one field using fl in query both will have same query time??

Re: Using solrcloud6 and solrj client with HAProxy.

On 9/1/2016 6:48 AM, Shawn Heisey wrote: > You can use the load balancer URL with HttpSolrClient instead of using > the zookeeper information with CloudSolrClient. At that point, you > will be dependent on haproxy to balance the load. It probably will not > react as fast to servers going down as

Re: Using solrcloud6 and solrj client with HAProxy.

On 9/1/2016 2:26 AM, Piyush Kunal wrote: > But I want to give priority to some nodes (with better configuration) > to have more load. Previously we used to use a HAproxy above all the > nodes which we can easily configure to put high loads on some nodes. > But if zookeeper is doing the

Replication Index fetch failed

Hi Replication seems to be in an endless loop. Anybody any idea? See below for logs. If you need more info, just let me know... INFO - 2016-09-01 14:30:42.563; [c:lvs s:shard1 r:core_node10 x:lvs_shard1_replica1] org.apache.solr.core.SolrDeletionPolicy; SolrDeletionPolicy.onCommit: commits:

Streaming expressions, delete

Hi, I've read up on the streaming expressions on the cwiki. The update decorator is ideal for quick imports from various sources such as jdbc, and combined with daemon it can be used for periodic delta imports too, which is very nice. But besides the update function, i would also expect a

Re: Default Field Cache

Are you looking for this ? org/apache/solr/core/SolrConfig.java:243 CacheConfig conf = CacheConfig.getConfig(this, "query/fieldValueCache"); if (conf == null) { Map args = new HashMap<>(); args.put(NAME, "fieldValueCache"); args.put("size", "1");

Re: Multiple rollups/facets in one streaming aggregation?

It is not possible to sort on multivalued fields out of the box ... ( indeed if you don't specify any logic, which value should considered as reference to sort ? ) Are you trying to sort or this happens by default ? Cheers On Thu, Sep 1, 2016 at 10:18 AM, subramani.new

Re: Multiple rollups/facets in one streaming aggregation?

Hello, I am exploring solr and its new feature ( Parallel Sql Interface and Stream api ). I have tried most of the api's and works fine. But, I am facing issue with multivalue field. My Json input has multi value fields. I trying to aggregate those fields but I am unable to. Exception : can

Can we get Filed List(fl) along with suggestions in Solr?

Hi, I am getting suggestions for solr using /suggest. I want some fields along with the suggestions also. How can I get that? Any suggestions. Thanks in advance Thanks and Regards M Pradeep Chandra

Using solrcloud6 and solrj client with HAProxy.

We are using a solrcloud 6.1 cluster with zookeeper. We have 6 nodes running behind the cluster. If I use solrj client with zookeeper, it will round robin across all the servers and distribute equal load across them. But I want to give priority to some nodes (with better configuration) to have