From b0e82bdaa9c47a0e4d32f7648b3af0bbc5e24ae1 Mon Sep 17 00:00:00 2001 From: Justin Littman Date: Mon, 2 Oct 2023 08:10:59 -0400 Subject: [PATCH] Fix db reset instructions. --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 1bb9ca0e..9bc9d1c2 100644 --- a/README.md +++ b/README.md @@ -318,7 +318,7 @@ You can view the raw markdown and copy the following checklist into a new issue - [ ] Delete all content under the `deposit` dirs and the `sdr2objects` dirs in each of the storage roots (`sdr2objects` is the configured "storage trunk", and `deposit` is where the bags that are used to build Moabs are placed). Storage root paths are listed in shared_configs for preservation_catalog (and should be the same as what's configured for presbots/techmd in the same env). So, e.g., `rm -rf /services-disk-stage/store2/deposit/*`, `rm -rf /services-disk-stage/store2/sdr2objects/*`, etc if the storage roots are `/services-disk-stage/store2/` etc. Deletions must be done from a preservation_robots machine for the env, as techMD and pres cat mount pres storage roots as read-only. - [ ] From a preservation_robots VM for the environment: Delete all content under preservation_robots' `Settings.transfer_object.from_dir` path, e.g. `rm -rf /dor/export/*` - [ ] preservation_catalog: _NOTE: this step likely not needed for most resets:_ for the environment being reset, `cap shared_configs:update` (to push the latest shared_configs, in case e.g. storage root or cloud endpoint locations have been updated) -- [ ] preservation_catalog: from any one host in the env to be reset, run `rake db:reset`. This should clear the DB, migrate to the current schema, and run the `rake db:seed` command (which should recreate the current storage root and cloud endpoint configs). See [this comment about DB reset approach for SDR resets](https://github.com/sul-dlss/argo/issues/4116#issuecomment-1688690735). +- [ ] preservation_catalog: from any one host in the env to be reset, perform a [db reset](https://github.com/sul-dlss/DeveloperPlaybook/blob/main/best-practices/db_reset.md) including seeding. - [ ] re-deploy preservation_catalog, preservation_robots, and techMD. This will bring the pres cat web services back online, bring the Sidekiq workers for all services back online, and pick up any shared_configs changes. Once the various SDR services are started back up, preserved content on storage roots and in the cloud will be rebuilt by regular accessioning. Since preservation just preserves whatever flows through accessioning, there's no need to worry about e.g. having particular APOs in place.