2de17bd396
deflake: TestDowngradeCancellationAfterDowngrading1InClusterOf3
...
Fixes: 65159a2b96fuweid89@gmail.com >
2025-02-22 08:22:17 -05:00
65159a2b96
*: Update cases related to Downgrade
...
1. Update DowngradeUpgradeMembersByID
If it's downgrading process, the desire version of cluster should be
target one.
If it's upgrading process, the desire version of cluster should be
determined by mininum binary version of members.
2. Remove AssertProcessLogs from DowngradeEnable
The log message "The server is ready to downgrade" appears only when the storage
version monitor detects a mismatch between the cluster and storage versions.
If traffic is insufficient to trigger a commit or if an auto-commit occurs right
after reading the storage version, the monitor may fail to update it, leading
to errors like:
```bash
"msg":"failed to update storage version","cluster-version":"3.6.0",
"error":"cannot detect storage schema version: missing confstate information"
```
Given this, we should remove the AssertProcessLogs statement.
Similar to #19313
Signed-off-by: Wei Fu <fuweid89@gmail.com >
2025-02-21 13:45:41 -05:00
1709422e21
Remove passing of anonymous visualize function
...
Signed-off-by: shashwat-jain <shashwat.jain@salesforce.com >
2025-02-12 11:26:10 +05:30
34546e9aa0
Add more debug info into waitTillSnapshot
...
Signed-off-by: Benjamin Wang <benjamin.ahrtr@gmail.com >
2025-02-07 12:14:49 +00:00
8866fce73f
test: update robustness doc and new case to reproduce 19179
...
NOTE: It's still hard to reproduce it. It may take 5 minutes or few
hours.
Signed-off-by: Wei Fu <fuweid89@gmail.com >
2025-02-06 10:24:52 -05:00
3b3243bbb9
increase timeout for MemberDowngradeUpgrade test
...
Signed-off-by: Gang Li <ganglica@google.com >
2025-02-05 23:58:46 +00:00
2bddcd4801
Merge pull request #19317 from siyuanfoundation/downgrade-robust-1
...
Remove some HealthInterval to reduce the time to run DowngradeUpgradeMembers
2025-02-03 10:22:08 +01:00
136dfbe5b5
*: introduce (*Op) Limit() interface for robustness
...
Since #19137 , kubernetes traffic profile is unable to send List request
with page size, because limit in option is not accessable. To fix it,
this fix is to introduce Limit() interface.
Fixes : #19292
Signed-off-by: Wei Fu <fuweid89@gmail.com >
2025-02-02 22:16:22 -05:00
6b0a9cd763
Remove some HealthInterval to reduce the time to run DowngradeUpgradeMembers.
...
Signed-off-by: Siyuan Zhang <sizhang@google.com >
2025-02-02 14:54:29 -08:00
e757a45d1d
Merge pull request #19269 from fuweid/test-robustness-list
...
tests/robustness: continue should ignore last key
2025-01-27 12:53:25 +01:00
f98fa31fa0
Merge pull request #19255 from AwesomePatrol/add-version-to-robustness-model
...
Add Version field to the robustness model
2025-01-27 12:07:12 +01:00
564e362408
tests/robustness: continue should ignore last key
...
Signed-off-by: Wei Fu <fuweid89@gmail.com >
2025-01-24 18:55:06 -05:00
e289ba3078
Merge pull request #19265 from redwrasse/redwrasse/rm-random-seed-robustness
...
Remove explicit random seed in robustness tests
2025-01-24 10:52:53 +01:00
43c6316420
robustness: remove explicit random seed in robustness tests.
...
Signed-off-by: redwrasse <mail@redwrasse.io >
2025-01-23 16:02:02 -08:00
b16b8dc6f3
migrate flag experimental-watch-progress-notify-interval to use watch-progress-notify-interval ( #19248 )
...
migrate flag experimental-watch-progress-notify-interval to use watch-progress-notify-interval
Signed-off-by: Gang Li <gangligit@gmail.com >
2025-01-23 21:08:30 +00:00
3dbe62fb02
Add Version field to the robustness model
...
It helps perform Txn before Compact in the same way
as it is done in Kubernetes.
Signed-off-by: Aleksander Mistewicz <amistewicz@google.com >
2025-01-23 14:11:23 +01:00
845a330e46
Implement Kubernetes like compaction
...
Signed-off-by: Marek Siarkowicz <siarkowicz@google.com >
2025-01-21 12:03:04 +01:00
5ccbeec769
Merge pull request #19218 from serathius/fix-selecting-experimental-flag
...
Fix passing compaction-batch-limit to etcd v3.4 and v3.5
2025-01-20 18:45:38 +01:00
8c989a1e37
Fix passing compaction-batch-limit to etcd v3.4 and v3.5
...
Signed-off-by: Marek Siarkowicz <siarkowicz@google.com >
2025-01-20 17:14:21 +01:00
10d7cea552
chore: enable early-return and superfluous-else from revive
...
Signed-off-by: Matthieu MOREL <matthieu.morel35@gmail.com >
Co-authored-by: Iván Valdés Castillo <iv@nvald.es >
2025-01-18 09:44:58 +01:00
9da01a8275
Merge pull request #19196 from gangli113/main
...
migrate flag experimental-compaction-batch-limit to use compaction-batch-limit
2025-01-17 07:43:22 +00:00
76aadeff38
Merge pull request #19169 from jmao-dd/jmao/robusttest-18089
...
Add Robustness test to reproduce issue 18089
2025-01-16 16:08:07 +01:00
7590a7ebae
tests: Reproduce #18089 in robustness tests
...
1) Use SleepBeforeSendWatchResponse failpoint to simulate slow watch
2) Decrease compact period from 200ms to 100ms to increase the probability of compacting on Delete
3) Introduce a new traffic pattern of 50/50 Put and Delete
With these three changes the `make test-robustness-issue18089` command can reproduce issue 18089.
Signed-off-by: Jiayin Mao <jiayin.mao@datadoghq.com >
2025-01-16 12:49:05 +00:00
ce4b4e533d
Merge pull request #19125 from siyuanfoundation/downgrade-robust-2
...
add MemberDowngradeUpgrade failpoint
2025-01-15 10:30:00 +01:00
8f51613574
add MemberDowngradeUpgrade failpoint
...
Signed-off-by: Siyuan Zhang <sizhang@google.com >
2025-01-15 01:06:39 +00:00
33d65fc90b
TestConfigFileDeprecatedOptions
...
Signed-off-by: Gang Li <gangligit@gmail.com >
2025-01-14 16:26:50 -08:00
d9d60be322
tests/robustness/traffic: should use rev=0 for create
...
Signed-off-by: Wei Fu <fuweid89@gmail.com >
2025-01-14 14:08:30 -05:00
abf65dde01
Merge pull request #19176 from jmao-dd/jmao/increase-limiter-burst
...
tests: use high burst value in limiter.
2025-01-13 10:20:55 +01:00
a03bdaabd2
Merge pull request #19175 from jmao-dd/jmao/fix-typo
...
tests: fix wrong number in comment.
2025-01-13 10:05:18 +01:00
47bd258fc8
tests: use high burst value in limiter.
...
Use the highest MaximalQPS of all traffic profiles as burst otherwise actual traffic may be accidentally limited.
For example, if burst is set to 200 it is unlikely traffic can achieve higher than 200.
Signed-off-by: Jiayin Mao <jiayin.mao@datadoghq.com >
2025-01-11 21:15:32 +00:00
5702d87501
tests: fix wrong number in comment.
...
The expected sum of weights should be 100, not 1000.
Signed-off-by: Jiayin Mao <jiayin.mao@datadoghq.com >
2025-01-11 21:07:44 +00:00
08e4d6d9c2
robustness: only run MemberDowngrade test for high SnapshotCatchUpEntries.
...
Signed-off-by: Siyuan Zhang <sizhang@google.com >
2025-01-09 16:12:02 -08:00
3d562c357f
Merge pull request #19137 from AwesomePatrol/use-new-interface-in-robustness-tests
...
Use new interface in robustness tests
2025-01-09 11:27:20 +01:00
00e5b654a1
Merge pull request #19095 from ahrtr/wal_20241221
...
Still return continuous WAL entries when running into ErrSliceOutOfRange
2025-01-08 14:14:19 +00:00
8335e70304
Migrate all OldTxn calls to Txn
...
Signed-off-by: Aleksander Mistewicz <amistewicz@google.com >
2025-01-08 11:48:26 +01:00
7c7d3ce8ae
Migrate all OldGet calls to Get
...
Signed-off-by: Aleksander Mistewicz <amistewicz@google.com >
2025-01-08 11:48:26 +01:00
524afd20d1
Set GetOnFailure=true to better simulate k8s traffic
...
Signed-off-by: Aleksander Mistewicz <amistewicz@google.com >
2025-01-08 11:48:26 +01:00
3247bc42bb
Make traffic robustness test use kubernetes.Interface
...
Signed-off-by: Aleksander Mistewicz <amistewicz@google.com >
2025-01-08 11:48:26 +01:00
747ef5f5fe
Add MemberDowngrade failpoint
...
Signed-off-by: Siyuan Zhang <sizhang@google.com >
2024-12-23 13:08:13 -08:00
152de1fa7e
Still return continuous WAL entries when running into ErrSliceOutOfRange
...
Signed-off-by: Benjamin Wang <benjamin.ahrtr@gmail.com >
2024-12-21 15:33:14 +00:00
647f1621d6
fix: enable gosimple linter
...
Signed-off-by: Matthieu MOREL <matthieu.morel35@gmail.com >
2024-12-03 07:32:22 +01:00
f2472d4b80
Handle non-linearized MemberList in v3.4 for robustness tests
...
Signed-off-by: Marek Siarkowicz <siarkowicz@google.com >
2024-12-02 14:32:38 +01:00
75cd1369a5
fix: enable gofumpt instead of gofmt linter in tests
...
Signed-off-by: Matthieu MOREL <matthieu.morel35@gmail.com >
2024-11-28 21:08:42 +01:00
0fcf6e80b4
Merge pull request #18928 from serathius/robustness-qps-150
...
Reduce QPS requirement to 100
2024-11-24 23:16:46 -07:00
fe45307dd4
Merge pull request #18905 from serathius/robustness-duplicated-puts-3
...
Robustness duplicated puts 3
2024-11-24 11:20:29 +01:00
b85c6ba7b4
Reduce QPS requirement to 100
...
Signed-off-by: Marek Siarkowicz <siarkowicz@google.com >
2024-11-20 20:49:56 +01:00
14f4df4019
Merge pull request #18930 from serathius/robustness-jitter
...
Add jitter to failpoint injection to cover periodily executed compaction
2024-11-20 20:47:03 +01:00
d961c81813
Add jitter to failpoint injection to cover periodily executed compaction
...
Signed-off-by: Marek Siarkowicz <siarkowicz@google.com >
2024-11-20 19:58:01 +01:00
3d33c09c46
Multiply return time by 100 in tests to detect off by one differences
...
Signed-off-by: Marek Siarkowicz <siarkowicz@google.com >
2024-11-17 21:55:30 +01:00
668834b7df
Allow duplicated put requests
...
Signed-off-by: Marek Siarkowicz <siarkowicz@google.com >
2024-11-17 21:51:45 +01:00
