Conversation
|
recnums are deprecated and barely used. I would not add new code for it. |
|
@akshatsikarwar requested it, should I remove it? |
roborivers
left a comment
roborivers
left a comment
There was a problem hiding this comment.
Coding style check: Error. ⚠.
Smoke testing: Error ⚠.
Cbuild submission: Success ✓.
Regression testing: 4/513 tests failed ⚠.
The first 10 failing tests are:
sc_transactional
scindex
queuedb_rollover
dupewrite
roborivers
left a comment
roborivers
left a comment
There was a problem hiding this comment.
Coding style check: Error. ⚠.
Smoke testing: Error ⚠.
Cbuild submission: Error ⚠.
Regression testing: 11/513 tests failed ⚠.
The first 10 failing tests are:
osql_cleanup [setup failure]
renametable [setup failure]
selectv_rcode_serialize_reads_like_writes_generated [setup failure]
drop_tblnum [setup failure]
remsql [setup failure]
sc_truncate
sc_transactional
queuedb_rollover_noroll1_generated
dupewrite
lostwrite
roborivers
left a comment
roborivers
left a comment
There was a problem hiding this comment.
Coding style check: Error. ⚠.
Smoke testing: Error ⚠.
Cbuild submission: Error ⚠.
Regression testing: 5/513 tests failed ⚠.
The first 10 failing tests are:
sc_truncate
sc_truncate_lockorder_generated
scindex
triggersc_latency
dupewrite
|
/plugin-branch count_fix |
roborivers
left a comment
roborivers
left a comment
There was a problem hiding this comment.
Coding style check: Error. ⚠.
Smoke testing: Error ⚠.
Cbuild submission: Success ✓.
Regression testing: 2/511 tests failed ⚠.
The first 10 failing tests are:
queuedb_rollover
ddl_no_csc2
roborivers
left a comment
roborivers
left a comment
There was a problem hiding this comment.
Coding style check: Error. ⚠.
Smoke testing: Error ⚠.
Cbuild submission: Success ✓.
Regression testing: 6/511 tests failed ⚠.
The first 10 failing tests are:
replay_history
sc_truncate
queuedb_rollover_noroll1_generated
queuedb_rollover
prepare
ddl_no_csc2
roborivers
left a comment
roborivers
left a comment
There was a problem hiding this comment.
Coding style check: Error. ⚠.
Smoke testing: Error ⚠.
Cbuild submission: Success ✓.
Regression testing: 4/511 tests failed ⚠.
The first 10 failing tests are:
sc_truncate_lockorder_generated
queuedb_rollover_noroll1_generated
queuedb_rollover
ddl_no_csc2
Signed-off-by: Salil Chandra <schandra107@bloomberg.net>
Signed-off-by: Salil Chandra <schandra107@bloomberg.net>
3 participants
See commit by commit
For first commit:
bdb_count doesn't use dta if no indices are available, led to seg fault. Instead of calling bdb_count_int call bdb_direct_count_int which does use dta if no index available.
Wrote robomark test direct_count for this
For second commit:
bdb_count_int had an optimization to find count using recnums first. Port that over to bdb_direct_count_int
Ported code and cleaned up into functions from https://github.com/bloomberg/comdb2/blob/main/bdb/count.c#L66
DIFFERENCE: DB_REP_HANDLE_DEAD isn't set to BDB_ERR_DEADLOCK anymore. Don't think this would ever be a possible rc though since locks acquired before calling bdb_direct_count_int?