Question 1

Solr Traffic Bandwidth

Accepted Answer

Opensolr does not bill by the number of requests. It bills the bytes your index sends back. A tight query and a sloppy one cost the same to make, but not the same to answer. 1,000,000 lean responsesa few bytes each1 GBsame bandwidthsame bill1 heavy responseevery field, every row, no filtersonly the ...

Question 2

Solr Best Practices

Accepted Answer

Solr does not run out of memory because it is busy. It runs out because of what it was asked for: a page ten thousand results deep, a facet on a field that has to be un-inverted first, a query that returns every field of a thousand documents. This page is the list of the requests and settings that c...

Question 3

Solr 9 Features and What's New

Accepted Answer

Apache Solr 9 is the release that put vector search inside Solr itself, and the release that quietly deleted several classes a working Solr 8 configuration still names. Those two facts are the whole story: it unlocked everything Opensolr does with AI, and it is the reason a config file that booted f...

Question 4

Laravel Opensolr Search: Sample App with Scout, Vue 3 and Hybrid Search

Accepted Answer

A complete, working Laravel 13 application that puts a Vue 3 search interface in front of an Opensolr index. Search is hybrid: keyword relevance and semantic similarity fused per document, so "sleepy pets" finds the article about cats napping. One page searches an Eloquent model through Laravel Scou...

Question 5

How can I save disk space for my index?

Accepted Answer

A smaller index is mostly a schema decision. Each of the levers below trades a little query capability for a lot of disk, and you choose which ones you can afford to give up. 01 · The levers, and what each one costs you ChangeWhat you gainWhat you give upUse Point fields (pint, plong, pdate) instead...

Question 6

Where Solr Memory Goes - JVM Heap and Memory Mapped Files

Accepted Answer

Solr runs out of memory because of what it was asked for. Knowing where each request lands in memory is the first step to asking for less. Solr runs inside a JVM with a fixed maximum heap. Everything the JVM cannot fit there ends in an OutOfMemoryError, and a Java process that hits one is not reliab...

Question 7

Solr Performance - Ask for Fewer Rows and Fields

Accepted Answer

The cheapest Solr optimisation is asking for less: fewer rows, and only the fields the page shows. A search result page shows ten or twenty results. A request for a thousand costs a thousand documents read from disk, decompressed, cached and serialized, on every single call. The numbers below are on...

Question 8

Solr Deep Paging and cursorMark

Accepted Answer

Deep paging is the most expensive request a search page can receive, and the easiest one for a bot to send. Asking for start=500000&rows=10 does not skip half a million documents. Solr has to find, score and rank every one of them to know which ten come next, and it holds a queue of 500,010 entries ...

Question 9

Solr docValues and the TextField Trap

Accepted Answer

Faceting, sorting and grouping need a column of values per field. docValues keep that column on disk; without them Solr rebuilds it in the heap. Faceting, sorting, grouping and function queries all need the same thing: the value of a field for a given document. The inverted index answers the opposit...

Question 10

Solr Cache Sizing - filterCache queryResultCache documentCache

Accepted Answer

Every Solr cache trades heap for speed. Sized against your own document count, they help; copied from an example, they can take the heap. Solr caches are worth having, and they are also the easiest way to lose a heap. Four of them are configurable in solrconfig.xml, and each entry they hold is a liv...

Question 11

The Cost of Faceting Grouping and Highlighting in Solr

Accepted Answer

Facets, grouping and highlighting are what make a search page useful, and what make a query expensive. These three components do more work per request than the search itself, and all three scale with something other than the number of results returned. Never send facet.limit=-1 on a high cardinality...

Question 12

Solr Queries That Fail on Their Own

Accepted Answer

Some queries fail because of what they are, not because of the server they run on. These are the shapes to recognise. Some requests do not degrade — they throw. These are the ones we see most often in the error logs of indexes that were working fine the day before. What arrivesWhat happensWhat to do...

Question 13

CMS Integrations That Commit Every Second

Accepted Answer

Some CMS search modules send every update with commitWithin=1000: each one asks Solr to make the change searchable within a second. On a busy site that means a new searcher every second, all day.What each commit costsA commit writes a new small segment, opens a new searcher, warms its caches, and la...

Question 14

Inline Base64 Images in Indexed HTML

Accepted Answer

Many CMS editors let authors paste a picture straight into the text. The picture then lives inside the HTML as a data URI, and when the rendered page is sent to Solr, the picture is indexed together with the words around it.How to spot itA field value containing src="data:image/png;base64, (or image...

Question 15

When One Document Keeps Crashing Solr

Accepted Answer

If indexing fails at regular intervals, every time your CMS or import job runs, the cause is often a single document, not the server.Failed batches are sent againMost CMS integrations send documents in batches and mark the whole batch as not indexed when a request fails, then send it again on the ne...

Question 16

Suggesters That Rebuild on Every Commit

Accepted Answer

The Solr suggester keeps its own dictionary of suggestions. How and when that dictionary is rebuilt decides whether autocomplete costs nothing or keeps a CPU core busy all day.buildOnCommit rebuilds from the whole indexWith DocumentDictionaryFactory, a build reads the stored field of every document ...

Question 17

Dynamic Field Explosion in Solr

Accepted Answer

Dynamic fields let you add fields without touching the schema. Putting a site, an environment, a language or a customer into the field name turns one index into tens of thousands of fields, and Solr pays for every field in every segment.Each field costs memory per segmentField metadata, norms and te...

Question 18

Indexing into Solr Without a Memory Spike

Accepted Answer

Indexing problems are almost always commit problems. A commit per document is a new searcher per document. Send documents in batches, not one at a timeA few hundred to a few thousand documents per request is the working range. One request per document spends most of its time on connection and commit...

Question 19

Never Put Binary Data or Base64 Into Solr

Accepted Answer

Solr indexes text. Binary data sent as a field value (a picture, a PDF, an attachment, a serialized object, encoded as base64) is the fastest way to make an index huge, slow and unstable. It is never searched, and it is paid for on every request.Every base64 string is a pile of unique termsBase64 is...

Question 20

N-gram Fields Belong on Short Text Only

Accepted Answer

An n-gram field indexes every substring of every word. It makes partial matching easy, and it multiplies the size of whatever it is applied to. Used on the wrong field, it is the most common cause of slow indexing and memory spikes.What an n-gram filter doesWith NGramFilterFactory minGramSize="2" ma...

Question 21

Solr 9 Dense Vectors and the knn Query Parser

Accepted Answer

Solr 9.0 is the release that let Solr match documents by meaning, not only by the words they contain. Two additions do it: a field type that stores vectors and a query parser that finds the nearest ones. Solr 9.0 introduced the DenseVectorField field type and the !knn query parser. Before that relea...

Question 22

Everything Else Solr 9 Added

Accepted Answer

Vectors are the headline of Solr 9, but not the only change. These are the other additions, and the ones that change how you run or call Solr. Java 11 is the floor, Java 17 the recommendationSolr 9 requires Java 11 as a minimum and is continuously tested against Java 11, 17 and 21. The reference gui...

Question 23

What Solr 9 Removed and What Only Moved to Modules

Accepted Answer

Solr 9 reorganised the old contribs into modules and removed a few components for good. Telling the two apart is what decides whether an old configuration needs a switch or a rewrite. Solr 9 reorganised the old contribs into modules: separate jars, in their own directories, switched on by name. A co...

Question 24

Solr 9 Defaults That Changed Under Your Configuration

Accepted Answer

A Solr 9 upgrade rarely breaks on a feature you use. It breaks on a default you never set. These are the ones that change behaviour under an unchanged configuration. A Solr 9 upgrade rarely fails on a feature you use. It fails on a default you never set and never thought about. These are the ones th...