roachtest: workaround for kv/splits failure#27756
Merged
craig[bot] merged 1 commit intocockroachdb:masterfrom Jul 19, 2018
Merged
roachtest: workaround for kv/splits failure#27756craig[bot] merged 1 commit intocockroachdb:masterfrom
craig[bot] merged 1 commit intocockroachdb:masterfrom
Conversation
The kv/splits/nodes=3 roachtest creates a 3 node cluster and then splits an empty table over and over until 500k ranges are created. Unfortunately, 2.1 uses slightly more memory per range than 2.0 and the result is that this test went from passing to failing in some bizarre ways (liveness failures and then apparent killing by some external system). The long term fix is to use less memory per range, probably by evicting idle ranges from memory completely, or replacing them with a minimal stub. In the short term, we can set the RocksDB cache size for the test lower to free up 3GB of memory for the test. Fixes cockroachdb#26081 Release note: None
Member
|
This change is |
Collaborator
Author
|
@spencerkimball I know you're looking at reducing the memory footprint for idle ranges. One thing I noticed during this investigation is the overhead for the |
Member
|
I'm targeting a deeper fix. I've seen |
spencerkimball
approved these changes
Jul 19, 2018
Collaborator
Author
|
bors r=spencerkimball |
craig bot
pushed a commit
that referenced
this pull request
Jul 19, 2018
27756: roachtest: workaround for kv/splits failure r=spencerkimball a=petermattis The kv/splits/nodes=3 roachtest creates a 3 node cluster and then splits an empty table over and over until 500k ranges are created. Unfortunately, 2.1 uses slightly more memory per range than 2.0 and the result is that this test went from passing to failing in some bizarre ways (liveness failures and then apparent killing by some external system). The long term fix is to use less memory per range, probably by evicting idle ranges from memory completely, or replacing them with a minimal stub. In the short term, we can set the RocksDB cache size for the test lower to free up 3GB of memory for the test. Fixes #26081 Release note: None Co-authored-by: Peter Mattis <petermattis@gmail.com>
Contributor
Build succeeded |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
The kv/splits/nodes=3 roachtest creates a 3 node cluster and then splits
an empty table over and over until 500k ranges are
created. Unfortunately, 2.1 uses slightly more memory per range than 2.0
and the result is that this test went from passing to failing in some
bizarre ways (liveness failures and then apparent killing by some
external system). The long term fix is to use less memory per range,
probably by evicting idle ranges from memory completely, or replacing
them with a minimal stub. In the short term, we can set the RocksDB
cache size for the test lower to free up 3GB of memory for the test.
Fixes #26081
Release note: None