Re: Solr & Kubernetes - how to configure the liveness

Thanks this is really interesting > > On 27 Oct 2021, at 18:36, Houston Putman wrote: > > Vincenzo, > > If you use the Solr Operator , it will > manage the upgrades for you in a safe manner (waiting for x number of > replicas to be healthy before moving ont

Re: Significant performance hit on replication compared to older version

The external file field feature works great. NFS mounts are the problem. That should be clear from my comment. When search indexes were on NFS, it was 100X slower. Search indexes, not external file field files. That feature wasn’t even in Solr when I accidentally ran on NFS. wunder Walter Underw

Re: Significant performance hit on replication compared to older version

Looks like the external file field feature certainly needs some love :) On Wed, 27 Oct 2021, 18:29 Walter Underwood, wrote: > So the “boost files” are files for an external file field? > > Ages ago I accidentally put search indexes on an NFS volume. Solr was 100X > slower. > > wunder > Walter U

Re: Solr & Kubernetes - how to configure the liveness

I think it will remain as 200 - it is returning the status of the cores. If the call itself fails then of course the HTTP status would reflect that. I think the Solr Admin UI uses this call on one of the cloud pages. Rob > On 27 Oct 2021, at 17:29, Vincenzo D'Amore wrote: > > HI Rob, thanks

Re: Is there a strict instruction hot to establish backups on S3 from Solr instance (8.10.1 version)

Hello Dmytro, In your solr.xml, you did not set the S3Repository to be the "default" BackupRepository, so you would need to use the "repository=s3" url param when interacting with the backup APIs. Also the S3Repository requires a "location" url Param as well, which you will need to set to either

Re: Solr & Kubernetes - how to configure the liveness

Vincenzo, If you use the Solr Operator , it will manage the upgrades for you in a safe manner (waiting for x number of replicas to be healthy before moving onto the next node). Hopefully the following documentation pages will help: - CRD Options for Update S

Re: Solr & Kubernetes - how to configure the liveness

HI Rob, thanks for your help. Do you know if in case of failure (initFailures not empty) /solr/admin/cores changes the http status code of the response in 500 (or everything that is not 200) ? On Wed, Oct 27, 2021 at 6:13 PM Robert Pearce wrote: > Take a look at the cores REST API, something lik

Re: Significant performance hit on replication compared to older version

So the “boost files” are files for an external file field? Ages ago I accidentally put search indexes on an NFS volume. Solr was 100X slower. wunder Walter Underwood wun...@wunderwood.org http://observer.wunderwood.org/ (my blog) > On Oct 27, 2021, at 7:37 AM, Dominic Humphries > wrote: > >

Re: Significant performance hit on replication compared to older version

Very good question, for which I currently have no answer! On Wed, 27 Oct 2021 at 17:15, Deepak Goel wrote: > Why wouldnt the performance hit not happen for 8.3.1? > > On Wed, 27 Oct 2021, 20:07 Dominic Humphries, > wrote: > > > At last, I think we've got it! > > > > Our external boost files liv

Re: Significant performance hit on replication compared to older version

Why wouldnt the performance hit not happen for 8.3.1? On Wed, 27 Oct 2021, 20:07 Dominic Humphries, wrote: > At last, I think we've got it! > > Our external boost files live on an NFS volume so they can be updated once > by a worker machine and all the followers will get the update. Which is all

Re: Solr & Kubernetes - how to configure the liveness

Take a look at the cores REST API, something like http://localhost:8983/solr/admin/cores?action=STATUS&wt=json Any failed cores will be in ‘initFailures’; cores which started will be under “status” Rob > On 27 Oct 2021, at 16:28, Vincenzo D'Amore wrote: > > Hi all, > > when a Solr instance

Re: Child doc question

Hi Folks, Wanted to follow up here. Can someone help me just answer whether what I'm hoping for is feasible? (a) The desired outcome is supported with the right model/queries (b) The desired outcome is not supported; maybe it could be in the future (c) The desired outcome is fundamentally unsuppo

Solr & Kubernetes - how to configure the liveness

Hi all, when a Solr instance is started I would be sure all the indexes present are up and running, in other words that the instance is healthy. The healthy status (aka liveness/readiness) is especially useful when a Kubernetes SolrCloud cluster has to be restarted for any configuration management

When using function query as a sort param in collapse parser, queryresultcache is not working

Hi all, I am using the below query params attached the screenshot. In collapse parser I am using the Functionquery param as $serviceable and in the function query I am using cluster_string_mv:c1. The issue is during queryresultcache key formation the $serviceable and cluster_string_mv:c1 are getti

Re: Significant performance hit on replication compared to older version

At last, I think we've got it! Our external boost files live on an NFS volume so they can be updated once by a worker machine and all the followers will get the update. Which is all very nice. But if we instead source those files from the local filesystem instead of one mounted from the network,

Re: Solr / OpenNLP integrations

Thanks, Alexandre. That's valuable feedback and I will work to improve the documentation. Thanks, Jeff On Tue, Oct 26, 2021 at 8:43 PM Alexandre Rafalovitch wrote: > I feel that it was a bit challenging to get the integration to work for > somebody who understood Solr but not as much of OpenNLP