Split dual-channel COB overrun tests to separate servers#1374
Conversation
1. The test isn't waiting long enough for the output buffer to overrun. This problem is happening because an error from the previous test is bleeding into the current test's logs. The simplest fix would be to split these tests. 2. Increased replication timeout to ensure sync fails due to output buffer overrun before a timeout occurs. Solve valkey-io#1367 Signed-off-by: naglera <anagler123@gmail.com>
|
Notice from the master test logs, the client which reached COB limit is client 11 which was created in |
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## unstable #1374 +/- ##
============================================
+ Coverage 70.59% 70.62% +0.02%
============================================
Files 117 117
Lines 63324 63324
============================================
+ Hits 44704 44722 +18
+ Misses 18620 18602 -18 |
|
is splitting the test really the only option? we can fetch the replica client ID and use it to search the log. I generally prefer we avoid spawning many servers if no real need |
zuiderkwast
left a comment
There was a problem hiding this comment.
Good job! This test case has been haunting us for a while.
To start many servers take some time, so it's good to avoid it to keep total test time shorter. But in this case maybe it's not a problem. I also agree that separating them is better for isolation of tests where we check for log messages.
@ranshid @xbasel Do you want to try to find other alternatives?
I also suffered this when dealing with #1297, i think split it is a good idea though it will increase the test time. |
) 1. The test isn't waiting long enough for the output buffer to overrun. This problem is happening because an error from the previous test is bleeding into the current test's logs. The simplest fix would be to split these tests. 2. Increased replication timeout to ensure sync fails due to output buffer overrun before a timeout occurs. Fixes valkey-io#1367 Signed-off-by: naglera <anagler123@gmail.com> (cherry picked from commit 7043ef0)
) 1. The test isn't waiting long enough for the output buffer to overrun. This problem is happening because an error from the previous test is bleeding into the current test's logs. The simplest fix would be to split these tests. 2. Increased replication timeout to ensure sync fails due to output buffer overrun before a timeout occurs. Fixes valkey-io#1367 Signed-off-by: naglera <anagler123@gmail.com> (cherry picked from commit 7043ef0) (cherry picked from commit c40dc4e)
) 1. The test isn't waiting long enough for the output buffer to overrun. This problem is happening because an error from the previous test is bleeding into the current test's logs. The simplest fix would be to split these tests. 2. Increased replication timeout to ensure sync fails due to output buffer overrun before a timeout occurs. Fixes valkey-io#1367 Signed-off-by: naglera <anagler123@gmail.com> (cherry picked from commit 7043ef0) (cherry picked from commit c40dc4e)
1. The test isn't waiting long enough for the output buffer to overrun. This problem is happening because an error from the previous test is bleeding into the current test's logs. The simplest fix would be to split these tests. 2. Increased replication timeout to ensure sync fails due to output buffer overrun before a timeout occurs. Fixes #1367 Signed-off-by: naglera <anagler123@gmail.com> (cherry picked from commit 7043ef0) (cherry picked from commit c40dc4e)
Fixes #1367