Google Git
Sign in
apache / lucene-solr / refs/heads/branch_8_10
commit6a6484ba396927727b16e5061384d3cd80d616b2[log] [tgz]
authorJason Gerlowski <gerlowskija@apache.org>Tue Oct 19 08:07:05 2021 -0400
committerJason Gerlowski <gerlowskija@apache.org>Fri Oct 22 10:06:10 2021 -0400
tree3731982d488557925d321aac1bc21c94f4312a79
parent68b0cdb554c53ca8567ac88b935c579b5586cda1 [diff]
SOLR-15696: Fix ShardBackupId parsing for backups

Incremental backups store metadata files whose names are constructed in
part on the name of individual shards. These filenames are then parsed
later on to extract the shard name and "backup ID".

The parsing in question was overly-brittle though, not accounting for
the variations that shard-splitting etc. produce in shard names. This
caused incremental backups to fail for any collection that had
previously performed a shard-split.

This commit updates the parsing logic to handle these cases more
flexibly, allowing post-shardsplit backups to now succeed.
3 files changed
tree: 3731982d488557925d321aac1bc21c94f4312a79
  1. .github/
  2. dev-docs/
  3. dev-tools/
  4. lucene/
  5. solr/
  6. .git-blame-ignore-revs
  7. .gitignore
  8. .hgignore
  9. build.xml
  10. README.md
README.md

Apache Lucene and Solr

Apache Lucene is a high-performance, full featured text search engine library written in Java.

Apache Solr is an enterprise search platform written using Apache Lucene. Major features include full-text search, index replication and sharding, and result faceting and highlighting.

Online Documentation

This README file only contains basic setup instructions. For more comprehensive documentation, visit:

Building Lucene/Solr

(You do not need to do this if you downloaded a pre-built package.)

Lucene and Solr are built using Apache Ant. To build Lucene and Solr, run:

ant compile

If you see an error about Ivy missing while invoking Ant (e.g., .ant/lib does not exist), run ant ivy-bootstrap and retry.

Sometimes you may face issues with Ivy (e.g., an incompletely downloaded artifact). Cleaning up the Ivy cache and retrying is a workaround for most of such issues:

rm -rf ~/.ivy2/cache

The Solr server can then be packaged and prepared for startup by running the following command from the solr/ directory:

ant server

Running Solr

After building Solr, the server can be started using the bin/solr control scripts. Solr can be run in either standalone or distributed (SolrCloud mode).

To run Solr in standalone mode, run the following command from the solr/ directory:

bin/solr start

To run Solr in SolrCloud mode, run the following command from the solr/ directory:

bin/solr start -c

The bin/solr control script allows heavy modification of the started Solr. Common options are described in some detail in solr/README.txt. For an exhaustive treatment of options, run bin/solr start -h from the solr/ directory.

Development/IDEs

Ant can be used to generate project files compatible with most common IDEs. Run the ant command corresponding to your IDE of choice before attempting to import Lucene/Solr.

  • Eclipse - ant eclipse (See this for details)
  • IntelliJ - ant idea (See this for details)
  • Netbeans - ant netbeans (See this for details)

Running Tests

The standard test suite can be run with the command:

ant test

Like Solr itself, the test-running can be customized or tailored in a number or ways. For an exhaustive discussion of the options available, run:

ant test-help

Contributing

Please review the Contributing to Solr Guide for information on contributing.

Discussion and Support