- What are the minimum requirements to run Geo?
- How does Geo know which projects to sync?
- Can I use Geo in a disaster recovery situation?
- What data is replicated to a secondary site?
- Can I
git pushto a secondary site?
- How long does it take to have a commit replicated to a secondary site?
- What if the SSH server runs at a different port?
- Is this possible to set up a Docker Registry for a secondary site that mirrors the one on the primary site?
- Can I login to a secondary site?
The requirements are listed on the index page
On each secondary site, there is a read-only replicated copy of the GitLab database. A secondary site also has a tracking database where it stores which projects have been synced. Geo compares the two databases to find projects that are not yet tracked.
At the start, this tracking database is empty, so Geo tries to update from every project that it can see in the GitLab database.
For each project to sync:
- Geo issues a
git fetch geo --mirrorto get the latest information from the primary site. If there are no changes, the sync is fast. Otherwise, it has to pull the latest commits.
- The secondary site updates the tracking database to store the fact that it has synced projects A, B, C, and so on.
- Repeat until all projects are synced.
When someone pushes a commit to the primary site, it generates an event in the GitLab database that the repository has changed. The secondary site sees this event, marks the project in question as dirty, and schedules the project to be resynced.
To ensure that problems with pipelines (for example, syncs failing too many times or jobs being lost) don’t permanently stop projects syncing, Geo also periodically checks the tracking database for projects that are marked as dirty. This check happens when
the number of concurrent syncs falls below
repos_max_capacity and there are no new projects waiting to be synced.
Geo also has a checksum feature which runs a SHA256 sum across all the Git references to the SHA values. If the refs don’t match between the primary site and the secondary site, then the secondary site will mark that project as dirty and try to resync it. So even if we have an outdated tracking database, the validation should activate and find discrepancies in the repository state and resync.
Yes, but there are limitations to what we replicate (see What data is replicated to a secondary site?).
Read the documentation for Disaster Recovery.
We currently replicate project repositories, LFS objects, generated attachments and avatars, and the whole database. This means user accounts, issues, merge requests, groups, project data, and so on, will be available for query.
Yes! Pushing directly to a secondary site (for both HTTP and SSH, including Git LFS) was introduced in GitLab 11.3.
All replication operations are asynchronous and are queued to be dispatched. Therefore, it depends on a lot of factors including the amount of traffic, how big your commit is, the connectivity between your sites, your hardware, and so on.
That’s totally fine. We use HTTP(s) to fetch repository changes from the primary site to all secondary sites.
Is this possible to set up a Docker Registry for a secondary site that mirrors the one on the primary site?
Yes. See Docker Registry for a secondary site.
Yes, but secondary sites receive all authentication data (like user accounts and logins) from the primary instance. This means you are re-directed to the primary for authentication and then routed back.