You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As I understand it there is a race condition between mounting a volume over SOLR and lagoon/lando/pygmy which can be addressed by recreating the core ( #97 )
The recommended way for getting a SOLR-8 running via @tobybellwood was something similar to:
FROM uselagoon/solr-8-drupal
CMD solr-recreate drupal /solr-conf && solr-foreground
This works great on lagoon
But when working locally in a tool like lando, the command gets overridden to
This can be worked around by setting up the cmd again in the .lando.yml file as an override for the service, but it feels dirty writing this multiple times in a project.
Which got me thinking, what if its moved into an entrypoint?
FROM uselagoon/solr-8-drupal
USER root
RUN echo -e "#!/bin/bash\nsolr-recreate drupal /solr-conf" > /lagoon/entrypoints/90-solr.sh
USER solr
CMD ["solr-foreground"]
So the issue request is...
Would it be possible to bake in an entrypoint for solr-8 and solr-8-drupal that runs the recreate logic.
Is there any negatives to it?
Could it use an ENV inside the entrypoint for the core name?
The text was updated successfully, but these errors were encountered:
As a followup to this @tobybellwood we're using this approach for Solr 8 on our site (worked fine in testing for a few weeks, just deployed to production and seems to be going fine).
We have three Solr cores (one with 2 indexes), and I used the same approach as above, echo -ing the solr-recreate command for each core into /lagoon/entrypoints/90-solr-recreate.sh.
As I understand it there is a race condition between mounting a volume over SOLR and lagoon/lando/pygmy which can be addressed by recreating the core ( #97 )
The recommended way for getting a SOLR-8 running via @tobybellwood was something similar to:
This works great on lagoon
But when working locally in a tool like lando, the command gets overridden to
This can be worked around by setting up the cmd again in the .lando.yml file as an override for the service, but it feels dirty writing this multiple times in a project.
Which got me thinking, what if its moved into an entrypoint?
So the issue request is...
The text was updated successfully, but these errors were encountered: