sql: CockroachDB crashes when executing the PoC query. #91301
Assignees
Labels
A-schema-changes
branch-master
Failures and bugs on the master branch.
branch-release-22.2
Used to mark GA and release blockers, technical advisories, and bugs for 22.2
C-bug
Code not up to spec/doc, specs & docs deemed correct. Solution expected to change code/behavior.
GA-blocker
O-community
Originated from the community
T-sql-foundations
SQL Foundations Team (formerly SQL Schema + SQL Sessions)
X-blathers-triaged
blathers was able to find an owner
Comments
SteveLeungYL
added
the
C-bug
|
Hello, I am Blathers. I am here to help you get the issue triaged. Hoot - a bug! Though bugs are the bane of my existence, rest assured the wretched thing will get the best of care here. I have CC'd a few people who may be able to assist you:
If we have not gotten back to your issue within a few business days, you can try the following:
🦉 Hoot! I am a Blathers, a bot for CockroachDB. My owner is otan. |
blathers-crl
bot
added
A-cdc
|
cc @cockroachdb/cdc |
HonoreDB
added
A-schema-changes
T-sql-schema-deprecated
|
This sounds a lot like #90306. |
|
It's in fact a shocking bug in the declarative schema changer. I'm tagging it as a GA-blocker, the fix should be simple, low-impact, and the severity justifies not letting this bug fly in 22.2.0. |
postamar
added
GA-blocker
branch-release-22.2
This was referenced Nov 5, 2022
craig bot
pushed a commit
that referenced
this issue
Nov 8, 2022
91100: logictest: only run create_as_non_metamorphic under local r=ajwerner a=ajwerner This test is very expensive because it was writing 100Mib of data. It now writes more like 10. Nevertheless, let's not run it in so many places. I did verify that it still tests what it was intended to test. Fixes #91080 Release note: None 91253: ui: fiz size chart when first load r=maryliag a=maryliag Previously, the chart sizes were being properly updated when resized, but the initial value was not correct. This commit adds the call to handle resize so it can load the correct size on first load. Fix #91207 https://www.loom.com/share/3faf00b251e849d59120ce0b6e94047f Release note: None 91283: ui: add horizontal scroll txn insight details r=maryliag a=maryliag Previously, the table of "waiten on" inside the Transaction Insights details didn't have a horizontal scroll, not letting the user to see the full page on CC console. This commits adds the proper scroll to it. Fixes #91199 Before https://www.loom.com/share/80437381cd4546cfad9692c7718de38c After https://www.loom.com/share/7363d53bec1d49a0be1b90ab0c9069f2 Release note (bug fix): Add horizontal scroll on "waited on" table on Transaction Insight details page. 91307: scbuild: error on duplicate columns in ALTER PRIMARY KEY r=ajwerner a=postamar This check has been missing from the declarative schema changer builder. This would result in execution errors in later stages when the new index makes it into the table descriptor. Since these are internal errors, the user experience is correspondingly poor. This patch fixes this. Fixes #91301 Release note (bug fix): fixed a bug present only in earlier 22.2 release candidates, in which an ALTER PRIMARY KEY USING COLUMNS (x, x) statement would result in an internal error instead of the expected user-facing error with a pg-code. 91443: colexecop: remove stale TODO r=yuzefovich a=yuzefovich This commit removes now-stale TODO about `Closer.Close` being safe to execute even after `Release` call. The root cause was the intertwining of planning and execution infrastructure and has been recently addressed in #89052. Epic: None Release note: None 91446: colexec: protect columnarizer when closing not started input r=yuzefovich a=yuzefovich This commit makes sure that the columnarizer calls `InternalClose` only if it has been initialized. Previously, if `Columnarizer.Init` wasn't performed (most likely due to a panic in `Init` of another operator), the columnarizer's input would not be started, so when `InternalClose` called `input.ConsumerClosed`, that could lead to a nil pointer panic since `input.Ctx` would be `nil` if the input tried to do some logging (some processors do that). We now protect against this by short-circuiting `InternalClose` call altogether, similar to what we do in `Columnarizer.DrainMeta`. This makes it so that the columnarizer satisfies `Closer.Close` contract properly. Fixes: #84902. Release note: None 91450: workload/schemachange: address intermittent failures with select stmts r=fqazi a=fqazi Fixes: #91445 Previously, we added logic to detect memory budget, and disk budget errors but those were only when opening a result set. Unfortunately, these errors can also be hit when consuming the rows themselves. To address this, this patch will detect the same errors and ignore them if detected consuming the result set. Release note: None Co-authored-by: Andrew Werner <[email protected]> Co-authored-by: maryliag <[email protected]> Co-authored-by: Marius Posta <[email protected]> Co-authored-by: Yahor Yuzefovich <[email protected]> Co-authored-by: Faizan Qazi <[email protected]>
postamar
pushed a commit
that referenced
this issue
Nov 8, 2022
This check has been missing from the declarative schema changer builder. This would result in execution errors in later stages when the new index makes it into the table descriptor. Since these are internal errors, the user experience is correspondingly poor. This patch fixes this. Fixes #91301 Release note (bug fix): fixed a bug present only in earlier 22.2 release candidates, in which an ALTER PRIMARY KEY USING COLUMNS (x, x) statement would result in an internal error instead of the expected user-facing error with a pg-code.
postamar
pushed a commit
that referenced
this issue
Nov 8, 2022
This check has been missing from the declarative schema changer builder. This would result in execution errors in later stages when the new index makes it into the table descriptor. Since these are internal errors, the user experience is correspondingly poor. This patch fixes this. Fixes #91301 Release note (bug fix): fixed a bug present only in earlier 22.2 release candidates, in which an ALTER PRIMARY KEY USING COLUMNS (x, x) statement would result in an internal error instead of the expected user-facing error with a pg-code.
exalate-issue-sync
bot
added
T-sql-foundations
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Describe the problem
The latest version of CockroachDB (e01f162) crashes when executing the following query:
To Reproduce
Here is the detail steps to reproduce the bug.
Expected behavior
The query should throw an error indicating that the
USING COLUMNS (c2, c2)might contain a semantic error.Additional data / screenshots
Stack Trace:
Environment:
Additional context
CockroachDB crashes.
Jira issue: CRDB-21196
The text was updated successfully, but these errors were encountered: