-
Notifications
You must be signed in to change notification settings - Fork 1.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
A confusing warning for a keyspace with Network-Topology: "SimpleStrategy replication class is not recommended, but was used for keyspace scylla_bench" #21084
Comments
The issue is not reproduced in a standard master longevity (like 100gb-4h), but it is reproduced on master with this scenario of creating a new keyspace (scylla_bench) with initial-tablets(1):
PackagesScylla version: Kernel Version: Issue description
Describe your issue in detail and steps it took to produce it. ImpactDescribe the impact this issue causes to the user. How frequently does it reproduce?Describe the frequency with how this issue can be reproduced. Installation detailsCluster size: 3 nodes (i4i.large) Scylla Nodes used in this run:
OS / Image: Test: Logs and commands
Logs:
|
@ptrsmrn could this be related to the recent changes you introduced? |
I'm on master OSS on this commit 86fd9ce, which has 2d9b8f2, and if I create KS like that, it doesn't return this warning:
|
Perhaps |
I see a couple of things happened just before this warning got logged, e.g.:
If I do truncate on a table belonging to this ks in cqlsh, I don't get this warning (neither in cqlsh nor in scylla server). I think something is internally recreating this ks and either sets an invalid replication strategy, or it is corrected only later. |
Anyways, I don't think it's a major thing. |
Thanks @ptrsmrn , so according to this, a reproduction should be quite simple:
by the way, the logs may indeed point the warnings come right after scylla-bench start running -
So according to this, the warning message is wrong. It would better, of course, be something like "keyspace already exist with network-strategy" or no warning at all. |
|
Yes and no, because SimpleStrategy doesn't work with tablets, with which the orignal ks was created, so when I only change the replication strategies I get an error, not a warning:
But, if instead one tries to create a ks with "if not exists", then all the validation is triggered and we may get these warnings, e.g.:
That being said, I'm not sure if these warnings are all that bad, after all someone wants to create a ks with not recommended options, which we are warning about. The query is succesful in the sense that it doesn't return an error (nor creates a ks). |
@bhalevy , btw, it looks like the source of the issue comes from this line: |
@yarongilor definitely. SimpleStrategy is deprecated and we should stop using it everywhere. |
Packages
Scylla version:
6.3.0~dev-1.tabletmerge.20241013.259dcf894703
with build-id#NO_BUILDID
Kernel Version:
6.8.0-1016-aws
Issue description
Describe your issue in detail and steps it took to produce it.
in a master latest based build ( reproduced in 6.3 master as well) a keyspace with a network-strategy is created, as reported in node-1,
yet later on, the nodes report a warning that it has a simple-strategy.
SCT issues a create-keyspace cql like:
[node-1] keyspace creation:
[node-3] warning messages:
then, 20 minutes later, node-1 also got these warning:
node-1 does show the keyspace is created with tablets like:
The table of
system_schema.scylla_keyspaces
has:Some tablets activity in logs is:
Impact
Describe the impact this issue causes to the user.
How frequently does it reproduce?
Describe the frequency with how this issue can be reproduced.
Installation details
Cluster size: 3 nodes (i3.large)
Scylla Nodes used in this run:
OS / Image:
ami-07af88cdc0b3d8af9
(aws: undefined_region)Test:
byo-longevity-test-yg
Test id:
ecba0625-b97d-4d41-b708-7538ce55fcca
Test name:
scylla-staging/yarongilor/byo-longevity-test-yg
Test method:
tablets_split_merge_test.TabletsSplitMergeTest.test_tablets_split_merge
Test config file(s):
Logs and commands
$ hydra investigate show-monitor ecba0625-b97d-4d41-b708-7538ce55fcca
$ hydra investigate show-logs ecba0625-b97d-4d41-b708-7538ce55fcca
Logs:
Jenkins job URL
Argus
The text was updated successfully, but these errors were encountered: