Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/xenit-eu/docker-solr
Xenit specific Alfresco Solr images
https://github.com/xenit-eu/docker-solr
alfresco docker solr
Last synced: 21 days ago
JSON representation
Xenit specific Alfresco Solr images
- Host: GitHub
- URL: https://github.com/xenit-eu/docker-solr
- Owner: xenit-eu
- License: apache-2.0
- Created: 2019-07-15T12:00:29.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2024-12-06T02:31:03.000Z (about 1 month ago)
- Last Synced: 2024-12-06T03:25:34.701Z (about 1 month ago)
- Topics: alfresco, docker, solr
- Language: Shell
- Size: 887 KB
- Stars: 0
- Watchers: 2
- Forks: 1
- Open Issues: 14
-
Metadata Files:
- Readme: README.md
- Changelog: Changelog.md
- License: LICENSE
Awesome Lists containing this project
README
# Solr in Docker
Alfresco-specific solr docker images, compatible with Alfresco versions higher than 4.2.
## Images created
* [`alfresco-solr6`] = solr6 images (alfresco search services), using jetty
* [`alfresco-solr6-xenit`] = solr6 images (alfresco search services) with [alfred-telemetry-solr](https://github.com/xenit-eu/alfred-telemetry) and [custom solr backup/restore](https://github.com/xenit-eu/solr-backup) installed
## Supported Tags
* [`:1.3.1`, `:1.4.0`] = version of the alfresco search services used for solr6
## Overview
This is the repository building Solr Docker images.
All images are automatically built by [jenkins-2](https://jenkins-2.xenit.eu) and published to [hub.xenit.eu](https://hub.xenit.eu).
Community images are built by a [github workflow](https://github.com/xenit-eu/docker-solr/actions) and published to [docker hub](https://hub.docker.com/u/xenit).
## Example docker-compose files
Each time images are built, tests are automatically ran on containers started via docker-compose files.
If sharding is involved, then a valid license file which allows clustering is needed (\*)
(\*) this is conform alfresco documentation, although in practice I could run a sharded setup also without a clustering license. The admin console does not have the Sharding screen, but indexing and search actually works.
## Environment variables
There are several environment variables available to tweak the behaviour. The variables are read by an init script which further replaces them in the relevant files. Such relevant files include:
* solrcore.properties
* shared.properties
* schema.xml
* server.xml (ports, ssl-related properties for tomcat-based images)
* solr.in.sh (ports, ssl-related properties, paths for jetty-based images)
`solrcore.properties` can be set via a generic mechanism by setting environment variables of the form GLOBAL_WORKSPACE\_\ for workspace store, GLOBAL_ARCHIVE\_<\parameter\> for archive store, GLOBAL\_\ for all cores.
`shared.properties` can be set via a generic mechanism by setting environment variables of the form SHARED\_\.
A subset of the properties have also dedicated environment variables e.g. ALFRESCO_ENABLE_TRACKING. Generic variables
take precedence.
See also environment variables from lower layers: [`docker-openjdk`](https://github.com/xenit-eu/docker-openjdk)
and [`docker-tomcat`](https://github.com/xenit-eu/docker-tomcat) and - for Xenit-specific
images - [`alfred-telemetry`](https://github.com/xenit-eu/alfred-telemetry/tree/master/alfred-telemetry-solr)
| Variable | solrcore.property variable | java variable | Default | Comments |
|--------------------------------------------|---------------------------------|---------------|----------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| TEMPLATE | alfresco.template | | rerank | |
| CORES_TO_TRACK | | | alfresco;archive | loop over values to create config folders
solr6 only |
| CORES_ALFRESCO | | | alfresco | in case of sharded setups, cores to be created on the current host, separated by ";"
Example: alfresco-01;alfresco-02
Leave default for non-sharded setup
solr6 only |
| SOLR_DATA_DIR | | | /opt/alfresco-search-services/data/index | solr6 only |
| SOLR_MODEL_DIR | | | /opt/alfresco-search-services/data/model | solr6 only |
| SOLR_CONTENT_DIR | | | /opt/alfresco-search-services/data/contentstore | solr6 only |
| SOLR_BACKUP_DIR | solr.backup.dir | | /opt/alfresco-search-services/data/solr6Backup | |
| ALFRESCO_SOLR_SUGGESTER_ENABLED | solr.suggester.enabled | | true | needs also changes to schema.xml, otherwise not correct. |
| ALFRESCO_SOLR_FACETABLE_CATEGORIES_ENABLED | | | | changes schema.xml |
| ALFRESCO_HOST | alfresco.host | | alfresco | |
| ALFRESCO_PORT | alfresco.port | | 8080 | |
| ALFRESCO_PORT_SSL | alfresco.port.ssl | | 8443 | |
| ALFRESCO_SSL | alfresco.secureComms | | https | changes also server.xml for tomcat-based images and solr.in.sh for jetty-based images |
| ALFRESCO_SECRET | alfresco.secureComms.secret | | none | to be set when ALFRESSCO_SSL is secret to setup the shared secret between solr and alfresco |
| ALFRESCO_ENABLE_TRACKING | enable.alfresco.tracking | | true | in the workspace core or shard |
| ALFRESCO_INDEX_CONTENT | alfresco.index.transformContent | | true | in the workspace core or shard |
| ALFRESCO_CORE_POOL_SIZE | alfresco.corePoolSize | | 8 | |
| ALFRESCO_DO_PERMISSION_CHECKS | alfresco.doPermissionChecks | | true | post filtering of results still happens on Alfresco side |
| ARCHIVE_ENABLE_TRACKING | enable.alfresco.tracking | | true | in the archive core |
| ARCHIVE_INDEX_CONTENT | alfresco.index.transformContent | | true | in the archive core |
| JETTY_PORT | | | 8080 | solr6 only |
| JETTY_PORT_SSL | | | 8443 | solr6 only |
| GLOBAL_WORKSPACE_\ | \ | | | for workspace core or shards |
| GLOBAL_ARCHIVE_\ | \ | | | for archive core |
| GLOBAL_ALL_\ | \ | | | for all cores |
| GLOBAL__\ | \ | | | for specific core |
| SHARED_\ | \ | | | for `shared.properties` (solr6 only) |
| \* SSL_KEY_STORE | | | ssl.repo.client.keystore for solr6, ssl.keystore for solr4 | |
| \* SSL_KEY_STORE_PASSWORD | | | kT9X6oe68t | |
| \* SSL_KEY_STORE_ALIAS | | | ssl.repo | alias of the repository cert in the keystore, to be used in the solr6 healthcheck script |
| \* SSL_TRUST_STORE | | | ssl.repo.client.truststore for solr6, ssl.truststore for solr4 | |
| \* SSL_TRUST_STORE_PASSWORD | | | kT9X6oe68t | |
| JSON_LOGGING | | | false | if true, sets the application logs in JSON format |
| ACCESS_LOGGING | | | false | if true, sets the access logs in JSON format |
| RESTORE_FROM_BACKUP | | | empty(false) | if true, starts the process of restoring from s3 backup |
\* = tested for solr6
## Docker-compose files
Besides the docker-compose files used in the tests, there are other example files in [src/main/resources](https://github.com/xenit-eu/docker-solr/src/master/src/main/resources/).
## Support & Collaboration
These images are updated via pull requests to the [xenit-eu/docker-solr](https://github.com/xenit-eu/docker-solr/) Github-repository.
## Monitoring
Xenit-specific variants of the images contain [alfred-telemetry-solr](https://github.com/xenit-eu/alfred-telemetry) resources necessary for monitoring. See [alfred-telemetry-solr](https://github.com/xenit-eu/alfred-telemetry) for more details.
Since some of the metrics rely on jmx, it is enabled by default in this image variant.
### [Jmxtrans-agent](https://github.com/jmxtrans/jmxtrans-agent/)
Can be used by including the following sections in the docker-compose-solr file (example below is for solr6):
```yaml
volumes:
- ./jmxtrans-agent:/jmxtrans
....
environment:
- JAVA_OPTS_jmxtrans=-javaagent:/jmxtrans/jmxtrans-agent-1.2.6.jar=/jmxtrans/jmxtrans-agent-solr6.xml
```
Example configuration files are in directory [jmxtrans-agent](src/integrationTest/resources/jmxtrans-agent/).
Update to the latest jar file for jmxtrans-agent.
The agent is not needed if alfred-telemetry is installed.
### [Jmx exporter](https://github.com/prometheus/jmx_exporter)
Currently there are no example configuration files for jmx exporter in this project.
The agent is not needed if alfred-telemetry is installed.
### How to build
To build a local version of the _solr_ image:
```bash
./gradlew buildDockerImage
```
To run the integration tests:
```bash
./gradlew integrationTests
```
To see all available tasks:
```bash
./gradlew tasks
```
If you have access to [Alfresco private repository](https://artifacts.alfresco.com/nexus/content/groups/private/) add the repository to build.gradle and add
```bash
-Penterprise
```
to your build command.
## Solr backup
In the case of a non-sharded setup, solr index is backed-up via a scheduled job in Alfresco.
Parameters for the backup (location, maximum number of backups to keep) are set on Alfresco's side and passed to solr via the scheduled job, which calls the replication handler from solr.
By default they are /opt/alfresco/alf_data/solr4Backup for solr4 and /opt/alfresco-search-services/data/solr6Backup for solr6.
In the case of a sharded setup, backup needs to be done manually.
A custom backup/restore mechanism to S3 has been included in xenit-specific images and is described [here](./Backup.md).
## FAQ
### How do I take a thread dump?
Either use:
* ```kill -3 1``` inside the container and then look at the logs of the container
* ```jstack -l 1``` inside the container. Make sure you are running the command as user 'solr'.