[MDEV-12897] unit.lf failed in buildbot Created: 2017-05-24  Updated: 2022-04-28  Resolved: 2022-04-28

Status: Closed
Project: MariaDB Server
Component/s: Tests
Affects Version/s: 10.0, 10.1, 10.2
Fix Version/s: N/A

Type: Bug Priority: Major
Reporter: Elena Stepanova Assignee: Unassigned
Resolution: Fixed Votes: 0
Labels: None
Environment:

kvm-fulltest --ps-protocol


Issue Links:
Duplicate
duplicates MDEV-15670 unit.my_atomic failed in buildbot wit... Closed
is duplicated by MDEV-14465 unit.lf failed in buildbot, sporadic Closed
Relates
relates to MDEV-27088 Server crash on ARM (WMM architecture... Closed
relates to MDEV-28430 lf_alloc isn't safe on aarch64 (or pp... In Review

 Description   

http://buildbot.askmonty.org/buildbot/builders/kvm-fulltest/builds/8688/steps/test_3/logs/stdio

unit.lf                                  w4 [ fail ]
        Test ended at 2017-05-08 08:42:15
 
CURRENT_TEST: unit.lf
# N CPUs: 4, atomic ops: gcc-atomics-smp
1..7
ok 1 - my_atomic_initialize() returned 0
# Testing lf_pinbox (with my_thread_init) with 30 threads, 3000 iterations... 
Bail out! Signal 11 thrown
# 7 tests planned,  0 failed,  1 was last executed
 
 
mysqltest failed with unexpected return code 255



 Comments   
Comment by Elena Stepanova [ 2017-10-26 ]

10.0 8d0448d5075fe822e37f81b740e83f03

unit.lf                                  w1 [ fail ]
        Test ended at 2017-09-21 18:16:15
 
CURRENT_TEST: unit.lf
# N CPUs: 4, atomic ops: gcc-atomics-smp
1..7
ok 1 - my_atomic_initialize() returned 0
# Testing lf_pinbox (with my_thread_init) with 30 threads, 3000 iterations... 
ok 2 - tested lf_pinbox (with my_thread_init) in 0.0290057 secs (0)
# Testing lf_alloc (with my_thread_init) with 30 threads, 3000 iterations... 
# 0 mallocs, 17 pins in stack
ok 3 - tested lf_alloc (with my_thread_init) in 0.0487313 secs (0)
# Testing lf_hash (with my_thread_init) with 30 threads, 300 iterations... 
Bail out! Signal 11 thrown
# 7 tests planned,  0 failed,  3 was last executed
 
 
mysqltest failed with unexpected return code 255

Comment by Elena Stepanova [ 2017-11-23 ]

http://buildbot.askmonty.org/buildbot/builders/kvm-fulltest/builds/11078/steps/mtr_nm/logs/stdio

unit.lf                                  w1 [ fail ]
        Test ended at 2017-11-21 20:07:58
 
CURRENT_TEST: unit.lf
# N CPUs: 4, atomic ops: gcc-atomics-smp
1..7
ok 1 - my_atomic_initialize() returned 0
# Testing lf_pinbox (with my_thread_init) with 30 threads, 3000 iterations... 
ok 2 - tested lf_pinbox (with my_thread_init) in 0.0358399 secs (0)
# Testing lf_alloc (with my_thread_init) with 30 threads, 3000 iterations... 
Bail out! Signal 11 thrown
# 7 tests planned,  0 failed,  2 was last executed
 
 
mysqltest failed with unexpected return code 255
 
 
 - saving '/mnt/buildbot/build/mariadb-10.1.30/mysql-test/var/1/log/unit.lf/' to '/mnt/buildbot/build/mariadb-10.1.30/mysql-test/var/log/unit.lf/'

Comment by Michael Widenius [ 2022-02-04 ]

Failed 2 in buildbot on builder aarch64-debian-sid

1..6

  1. Testing lf_pinbox (with my_thread_init) with 30 threads, 30000 iterations...
    ok 1 - tested lf_pinbox (with my_thread_init) in 0.292456 secs (0)
  2. Testing lf_alloc (with my_thread_init) with 30 threads, 30000 iterations...
    Bail out! Signal 11 thrown
  3. 6 tests planned, 0 failed, 1 was last executed
Comment by Daniel Black [ 2022-02-11 ]

MDEV-27088 adjusted the lf hash tests to have sufficient iterations to catch errors.

The ARM based discussion reached consensus on https://github.com/MariaDB/server/pull/1951

even at bottom of MDEV-27088 test failures where observed.

Comment by Daniel Black [ 2022-04-28 ]

MDEV-28430 continues the long history of aarch64 (and small amount of ppc64le) failures.

Looking at the history the kvm-{fulltest,fulltest2,asan} (regardless of ps-protocol, didn't impact unit tests) test failure on unit.lf stopped in early 2019, with the exception of aarch64 and ppc64le.

Generated at Thu Feb 08 08:01:20 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.