Enhanced Cassandra health checks #1
Draft
Conversation
This file contains hidden or 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
adutra
force-pushed
the
java2731
branch
10 times, most recently
from
f464c0a to
bb556e2
Compare
Author
|
The PR has been retrofitted to rely on JAVA-2742 (health diagnostics in the driver itself). |
Author
|
PR for JAVA-2742 can be found here: apache/cassandra-java-driver#1435. |
adutra
force-pushed
the
java2731
branch
2 times, most recently
from
0455b95 to
cfc5d04
Compare
Motivation: The current Cassandra health check simply attempts to query one node in the cluster, randomly chosen, and sets the status to DOWN if that node cannot be queried. This is admittedly misleading, since Cassandra clusters are usually resilient to the loss of many nodes, and can still continue to serve application queries even if many nodes are down. Modification: This commit replaces the current health check by the DataStax driver's SessionDiagnostics feature, which perform the following checks: - Topology: reports how many nodes were found in each datacenter and globally, and how many were up or down. - Token range availability: inspects the ring and detects if certain token ranges are unavailable for the configured keyspace, local datacenter and consistency level. Note that these checks do not require querying the cluster; they rely solely on metadata exposed by the driver and on Gossip events, and thus execute very fast. The status is set to DOWN if: 1. The entire cluster is down; or 2. There are unavailable token ranges. Result: The health report now accurately describes the cluster health by setting it status to DOWN when queries are likely to fail due to UnavailableException errors.
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.
Motivation:
The current Cassandra health check simply attempts to query one
node in the cluster, randomly chosen, and sets the status to
DOWN if that node cannot be queried.
This is admittedly misleading, since Cassandra clusters are usually
resilient to the loss of many nodes, and can still continue to
serve application queries even if many nodes are down.
Modification:
This commit replaces the current health check by the following
checks:
and globally, and how many were up or down.
certain token ranges are unavailable for the configured
keyspace, local datacenter and consistency level.
Note that these checks do not require querying the cluster; they
rely solely on metadata exposed by the driver and Gossip event,
and thus execute very fast.
The status is set to DOWN if:
Result:
The health report now accurately describes the cluster health
by setting it status to DOWN when queries are likely to
fail due to UnavailableException errors.