roachtest: tpccbench/nodes=9/cpu=4/chaos/partition failed [fix pending] #60812
Assignees
Labels
branch-master
Failures and bugs on the master branch.
C-test-failure
Broken test (automatically or manually discovered).
O-roachtest
O-robot
Originated from a bot.
release-blocker
Indicates a release-blocker. Use with branch-release-2x.x label to denote which branch is blocked.
Comments
cockroach-teamcity
added
branch-master
|
(roachtest).tpccbench/nodes=9/cpu=4/chaos/partition failed on master@64c4aef909f4382523cd9248341ca9f4448d841a: More
Artifacts: /tpccbench/nodes=9/cpu=4/chaos/partition
See this test on roachdash |
|
(roachtest).tpccbench/nodes=9/cpu=4/chaos/partition failed on master@bf9744bad5a416a4b06907f0f3dd42896f7342f3: More
Artifacts: /tpccbench/nodes=9/cpu=4/chaos/partition
See this test on roachdash |
|
(roachtest).tpccbench/nodes=9/cpu=4/chaos/partition failed on master@5cfd7e5553a3072a1490d392390dddf968844215: More
Artifacts: /tpccbench/nodes=9/cpu=4/chaos/partition
See this test on roachdash |
|
NPE in |
|
Looks like https://github.com/angelapwen/cockroach/blob/eab1ae3735986bace291d09ec64571c5e094ffd6/pkg/kv/kvserver/replica_command.go#L2919-L2931 returned a nil |
|
... which it's allowed to do: cockroach/pkg/kv/kvserver/allocator.go Line 710 in e924d91 |
|
@aayushshah15 it looks like you most recently touched most of the involved code. Could you take a look to see if this is just a matter of adding a nil check or indicative of something else? |
|
(roachtest).tpccbench/nodes=9/cpu=4/chaos/partition failed on master@ec011620c7cf299fdbb898db692b36454defc4a2: More
Artifacts: /tpccbench/nodes=9/cpu=4/chaos/partition
See this test on roachdash |
|
Same stack trace in this failure. |
Looks like we've always been ignoring to check if that returned |
|
Oh, looks like I was dreaming. Indeed, it looks like I removed a nil check that should've been there. Fixing now. |
aayushshah15
added a commit
to aayushshah15/cockroach
that referenced
this issue
Feb 24, 2021
e924d91 introduced a bug by spuriously removing a nil check over the result of a call to `allocateTargetFromList`. This commit re-adds the check. The bug could cause a panic when `AdminRelocateRange` was called by the `StoreRebalancer` or the `mergeQueue` if one (or more) of the stores that are supposed to receive a replica for a range become unfit for receiving the replica (due to balancing reasons / or shifting constraints) _between when rebalancing decision is made and when it's executed_. Resolves cockroachdb#60812 Release justification: fixes bug that causes a panic Release note: None
tbg
changed the title
|
(roachtest).tpccbench/nodes=9/cpu=4/chaos/partition failed on master@c7e088826bc079620dfd3b5ae75d1c15cd9cd16d: More
Artifacts: /tpccbench/nodes=9/cpu=4/chaos/partition See this test on roachdash |
craig bot
pushed a commit
that referenced
this issue
Feb 25, 2021
57183: importccl,sql: support importing non-public schemas from pgdump r=adityamaru a=adityamaru Previously, a PGDUMP import did not support any non-public schema statements. Now that CRDB has user defined schemas, bundle format IMPORTs need to be taught how to parse, create and cleanup schema related PGDUMP operations. Note this PR only adds support for `CREATE SCHEMA` and usage of the schema in `CREATE TABLE/SEQUENCE` PGDUMP statements. `ALTER SCHEMA` statements are ignored and support might be added in a follow up. Release justification: low risk, high reward. Import PGDUMP does not support user defined schemas. Release note (sql change): IMPORT PGDUMP can now import dump files with non-public schemas. 60922: sql,cli: add payloads_for_trace builtin r=knz a=angelapwen Possibly the final step to #55733. Resolves #58608 😄 Previously it was quite cumbersome to view all payloads for a given trace: we needed to join on the `node_inflight_trace_spans` vtable to filter for span IDs that match a trace ID, then apply the `payloads_for_span()` builtin to each span ID. This patch adds syntactic sugar to the above query. Instead of ``` WITH spans AS ( SELECT span_id FROM crdb_internal.node_inflight_trace_spans WHERE trace_id = $TRACE_ID) ) SELECT * FROM spans, LATERAL crdb_internal.payloads_for_span(spans.span_id); ``` we can now simply use: ``` crdb_internal.payloads_for_trace($TRACE_ID); ``` and achieve the same result. The patch also adds all payloads for all long-running spans to the `crdb_internal.node_inflight_trace_spans` table of the debug.zip file. Release note (sql change): Add `payloads_for_trace()` builtin so that all payloads attached to all spans for a given trace ID will be displayed, utilizing the `crdb_internal.payloads_for_span()` builtin under the hood. All payloads for long-running spans are also added to debug.zip in the `crdb_internal.node_inflight_trace_spans` table dump. Co-authored-by: Tobias Grieger <[email protected]> Release justification: This patch is safe for release because it adds syntactic sugar to an internal observability feature. 61094: kvserver: re-add spuriously removed nil check in `relocateOne` r=aayushshah15 a=aayushshah15 bce8317 introduced a bug by spuriously removing a nil check over the result of a call to `allocateTargetFromList`. This commit re-adds the check. The bug could cause a panic when `AdminRelocateRange` was called by the `StoreRebalancer` or the `mergeQueue` if one (or more) of the stores that are supposed to receive a replica for a range become unfit for receiving the replica (due to balancing reasons / or shifting constraints) _between when rebalancing decision is made and when it's executed_. Resolves #60812 Release justification: fixes bug that causes a panic Release note: None 61097: opt: use computed columns to improve FDs and remove uniqueness checks r=rytaft a=rytaft **opt: use computed columns to build functional dependencies** This commit updates `MakeTableFuncDep` so that it adds equivalencies or synthesized columns to the table FDs for each of the computed columns available in the metadata. This will be necessary to support removing uniqueness checks in some cases in a future commit. Release justification: This commit is a low risk, high benefit change to existing functionality. Release note (performance improvement): The optimizer now infers additional functional dependencies based on computed columns in tables. This may enable additional optimizations and lead to better query plans. **opt: remove uniqueness checks when uniqueness inferred through FDs** This commit removes uniqueness checks for columns that can be inferred to be unique through functional dependencies. This is relevant in particular for `REGIONAL BY ROW` tables with a computed region column that depends on the primary key. In this case, uniqueness checks are never needed on the primary key, since uniqueness is already guaranteed by the primary index. Fixes #57720 Release justification: This commit is a low-risk, high benefit update to new functionality. Release note (performance improvement): Removed uniqueness checks on the primary key for REGIONAL BY ROW tables with a computed region column that is a function of the primary key columns. Uniqueness checks are not necessary in this case since uniqueness can be suitably guaranteed by the primary index. Removing these checks improves performance of INSERT, UPDATE, and UPSERT statements. 61100: diagnostics: lock while populating hardware information r=andy-kimball a=andy-kimball The shirou/gopsutil/host library that we use to gather hardware information during diagnostics reporting is not multi-thread safe. As one example, it lazily initializes a global map the first time the Virtualization function is called, but takes no lock while doing so. Work around this limitation by taking our own lock. This code never triggered race conditions before, but is doing so after recent changes I made to the diagnostics reporting code. Previously, we were using a single background goroutine to do both diagnostics reporting and checking for updates. Now we are doing each of those on different goroutines, which triggers race detection. Fixes #61091 Release justification: fixes for high-priority or high-severity bugs in existing functionality Release note: None 61105: builtins: implement ST_MakePoint and ST_MakePointM r=otan,rafiss a=andyyang890 This patch implements the geometry builtins `ST_MakePoint` and `ST_MakePointM`. Resolves #60857. Resolves #60858. Resolves #60859. Release justification: low-risk update to new functionality Release note (sql change): The geometry builtins `ST_MakePoint` and `ST_MakePointM` have been implemented and provide a mechanism for easily creating new points. Co-authored-by: Aditya Maru <[email protected]> Co-authored-by: angelapwen <[email protected]> Co-authored-by: Aayush Shah <[email protected]> Co-authored-by: Rebecca Taft <[email protected]> Co-authored-by: Andrew Kimball <[email protected]> Co-authored-by: Andy Yang <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
(roachtest).tpccbench/nodes=9/cpu=4/chaos/partition failed on master@83e70ce84b740e27e721c3b73c38a4b8b515094a:
More
Artifacts: /tpccbench/nodes=9/cpu=4/chaos/partition
Related:
roachtest: tpccbench/nodes=9/cpu=4/chaos/partition failed #60662 roachtest: tpccbench/nodes=9/cpu=4/chaos/partition failed C-test-failure O-roachtest O-robot branch-60635 release-blocker
roachtest: tpccbench/nodes=9/cpu=4/chaos/partition failed #60420 roachtest: tpccbench/nodes=9/cpu=4/chaos/partition failed C-test-failure O-roachtest O-robot branch-release-20.1 release-blocker
roachtest: tpccbench/nodes=9/cpu=4/chaos/partition failed #60130 roachtest: tpccbench/nodes=9/cpu=4/chaos/partition failed C-test-failure O-roachtest O-robot branch-release-20.2 release-blocker
See this test on roachdash
powered by pkg/cmd/internal/issues
The text was updated successfully, but these errors were encountered: