Fix uninitialized value in segno calculation #831
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
better to squash |
yjhjstz
approved these changes
Jan 3, 2025
|
Change commit message plz? |
This patch fixes misleading error message when test binaries `pg_regress/pg_isolation2_regress` are executed with the newer version of `.pm` scripts. For example: test binaries are builded at previous gpdb version (previous commit) and `./configure` script is called again, so `src/test/regress/GPTest.pm` (which stores version of gpdb at `$VERSION` constant) file is generated with newer version. And run of these tests exits with next error message: "The program "gpdiff.pl" is needed by pg_regress but was not found in the same\n directory as "/.../src/test/regress/pg_regress". But the real problem is next: `GP_VERSION` from binaries doesn't equal `$VERSION` from `GPTest.pm` script.
Fixme: src/backend/gpopt/translate/CContextDXLToPlStmt.cpp: // GPDB_12_MERGE_FIXME: we haven't seen the scan id yet, this scan id is likely // for dynamic pruning. Action: Check not used elsewhere. Static partition elimination result now referenced by m_part_indexes. Check and redundant declaration removed.
Fixme: src/backend/gpopt/translate/CTranslatorDXLToPlStmt.cpp: // GPDB_12_MERGE_FIXME: this duplicates a check in ExecInitProjectSet Action: None. Function SanityCheckProjectSetTargetList share some code with ExecInitProjectSet. But refactoring isn't trivial and will involve changes to upstream .c files. This check will be removed after PR #14302 is merged. For now, remove the FIXME tag to avoid confusion.
Strip out segment number when comparing ICW diff
…h when MQ has no space (#14925) If foreign table has lots of data, and we only retrieve few tuples in retrieve conn, Sender will be waiting in WaitLatch due to MQ has no space. Add testcase to test sender can be notified when cursor is closed.
* This commit is to fix issue #14528. Now we usually use CREATE EXTERNAL TABLE syntax to create an external table in gpdb, which can set default value while missing some options, such as option `execute_on`. But if we use CREATE FOREIGN TABLE syntax to create an external table, it won't set the default value while missing option `execute_on`. In this case, missing option `execute_on` will cause gpdb crashing. (Issue #14528) In this commit, when using CREATE FOREIGN TABLE syntax to create an external table and missing option `execute_on` , we will set default value for option `execute_on`.
This is the negligence of PR #14733. As the title said, this PR is just trying to fix the failed unit test of test_unit_gpcheckresgroupimpl.py
make each external table options delimited by space. currently only csv and text fmtopts output is delimited by space. This commit makes all formatters align
Minor correction that we don't need to call pstrdup() when initializing the query string. exec_mpp_query() would fill in a dummy string if it is empty. Noticed this when testing some specific workloads and seeing overheads in memory allocator caused by this. Didn't dig more into the impact but let's just fix it anyway.
rename master to main in gpload.py **Signed-off-by:** Yongtao Huang <[email protected]>
rename master to coordinator in gpfdist **Signed-off-by:** Yongtao Huang <[email protected]>
icproxy processes communicate with each other with the public addr:port (see gp_interconnect_proxy_addresses), but no auth control here. So, any connection may connect in and send any byte, it may cause some unexpected exceptions (e.g. OOM). In the PR, introduces a magic number field in the ICProxyPkt and does a sanity check: drop the corresponding package if magic number doesn't match. This method does not attempt to solve all problems, but only as a simple solution to avoid internal misaccess, looks it's good enough.
Remove previous hack in KeepLogSeg that added a case to deal with a (badly represented) invalid segment number. This was added for the sake of GetWALAvailability. But it's not needed if in that function we initialize the segment number to be retreated to the currently being written segment, so do that instead. Per valgrind-running buildfarm member skink, and some sparc64 animals. Discussion: https://postgr.es/m/[email protected] (cherry picked from commit 794e8e32bb5ae1b38a90cbae2a88895633797599)
done |
my-ship-it
approved these changes
Jan 6, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #ISSUE_Number
What does this PR do?
Type of Change
Breaking Changes
Test Plan
make installcheckmake -C src/test installcheck-cbdb-parallelImpact
Performance:
User-facing changes:
Dependencies:
Checklist
Additional Context
CI Skip Instructions