Tolerate a non-primary shard being down during startup #2727
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
lutter
force-pushed
the
lutter/startup-failed-shard
branch
from
7708109
to
2c9f35e
Compare
tilacog
approved these changes
Aug 27, 2021
Unfortunately, we can't just slap a "set port 'NNNN'" into the 'alter server' statement since we have installations that have a server without a port, and for those we first need to "add port 'NNNN'".
When we initialize a connection pool, we should only assume that the database for that pool is up and running. Previously, we used `import foreign schema` to create fdw tables locally, but that requires that Postgres access the remote server, and will therefore not work when that server is not up. We now create these tables by assuming the corresponding table on the remote server has the same schema as our local table, which avoids accessing the remote server. It also removes a subtle race where we could have pulled in the schema from a remote server before that ran its latest migrations.
This helps make it clearer how the underlying Diesel pool is accessed
When one of the configured shards is not available, defer setting it up (e.g., running migrations) until it becomes available.
lutter
force-pushed
the
lutter/startup-failed-shard
branch
from
2c9f35e
to
6775ad9
Compare
20 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
These changes make it so that if a shard that is not the primary is not available (down) when
graph-nodestarts up,graph-nodewill still start and respond to queries against subgraphs in the other shards. Furthermore, while the shard is down,graph-nodenow responds to queries against the failed shard with an error much quicker (roughly withinGRAPH_STORE_CONNECTION_TIMEOUT) where before it held on to the query indefinitely.Scenarios I tested when a non-primary shard is down:
graph-nodestarts up successfully