[MDEV-22082] Test spider/bugfix.wait_timeout occasionally fails with result difference Created: 2020-03-30  Updated: 2021-07-26  Resolved: 2021-07-26

Status: Closed
Project: MariaDB Server
Component/s: Storage Engine - Spider, Tests
Affects Version/s: 10.4, 10.5, 10.6
Fix Version/s: 10.4.21, 10.5.12, 10.6.4

Type: Bug Priority: Major
Reporter: Marko Mäkelä Assignee: Nayuta Yanagisawa (Inactive)
Resolution: Fixed Votes: 0
Labels: sporaric

Issue Links:
Duplicate
duplicates MDEV-26170 spider/bugfix.wait_timeout fails in b... Closed

 Description   

Note: I do not normally compile or test Spider. This time I did, on 10.4. The test failed in a 32-bit RelWithDebInfo build. It turns out that the test is nondeterministic, because I also got it to fail on a 64-bit Debug build, with enough retries:

./mtr --parallel=auto --repeat=100 spider/bugfix.wait_timeout{,,,,,,,,,,,}

Here are two sample failures:

CURRENT_TEST: spider/bugfix.wait_timeout
 
--- /mariadb/10.4/storage/spider/mysql-test/spider/bugfix/r/wait_timeout.result	2020-02-05 20:09:57.718285062 +0200
 
+++ /mariadb/10.4/storage/spider/mysql-test/spider/bugfix/r/wait_timeout.reject	2020-03-30 15:02:30.323127167 +0300
 
@@ -87,6 +87,21 @@
 
 25
 
 27
 
 29
 
+31
 
+33
 
+35
 
+37
 
+39
 
+41
 
+43
 
+45
 
+47
 
+49
 
+51
 
+53
 
+55
 
+57
 
+59
 
 connection child2_2;
 
 SELECT pkey FROM tbl_a ORDER BY pkey;
 
 pkey
 
@@ -105,6 +120,21 @@
 
 24
 
 26
 
 28
 
+30
 
+32
 
+34
 
+36
 
+38
 
+40
 
+42
 
+44
 
+46
 
+48
 
+50
 
+52
 
+54
 
+56
 
+58
 
 connection master_1;
 
 
 deinit

and from the 64-bit Debug build:

spider/bugfix.wait_timeout               w3 [ 3 fail ]
 
        Test ended at 2020-03-30 15:04:30
 
 
 
CURRENT_TEST: spider/bugfix.wait_timeout
 
--- /mariadb/10.4/storage/spider/mysql-test/spider/bugfix/r/wait_timeout.result2020-02-05 20:09:57.718285062 +0200
 
+++ /mariadb/10.4/storage/spider/mysql-test/spider/bugfix/r/wait_timeout.reject2020-03-30 15:04:29.948555836 +0300
 
@@ -105,6 +105,21 @@
 
 24
 
 26
 
 28
 
+30
 
+32
 
+34
 
+36
 
+38
 
+40
 
+42
 
+44
 
+46
 
+48
 
+50
 
+52
 
+54
 
+56
 
+58
 
 connection master_1;
 
 
 deinit
 
 
 
mysqltest: Result length mismatch

Please try to fix this in the earliest applicable version. I did not check any other source code revisions than 10.4e2f1f88fa60680cb87833a7cceb172f5d436a804 .

 Comments   
Comment by Alice Sherepa [ 2021-06-24 ]

10.4 09b03ff31b8fd71aee26 http://buildbot.askmonty.org/buildbot/builders/kvm-deb-groovy-amd64/builds/1323/steps/mtr/logs/stdio
10.5 60ed4797112533a88 http://buildbot.askmonty.org/buildbot/builders/kvm-deb-sid-amd64/builds/12292/steps/mtr/logs/stdio

Comment by Nayuta Yanagisawa (Inactive) [ 2021-07-20 ]

On 10.4 HEAD (3e5a11e) , I cannot reproduce the bug in my development environment (Vagrant box bento/ubuntu-20.04). I will look at it into further.

Comment by Nayuta Yanagisawa (Inactive) [ 2021-07-25 ]

I reproduced the failure. As serg pointed out in MDEV-26045, spider/bugfix.wait_timeout looks irrelevant to Spider's functionalities. Thus, I think that we could just disable the test until MDEV-26045 is closed.

Comment by Nayuta Yanagisawa (Inactive) [ 2021-07-25 ]

serg As I commented above, I think we could just disable the test temporarily. Please review: https://github.com/MariaDB/server/commit/f0b0d16d7ba9d36b13e04c7b90506f6ae5a7d79a

Comment by Sergei Golubchik [ 2021-07-26 ]

This is fine.

You might prefer a shorter comment in disabled.def, so that it wouldn't break the formatting as in

parts.partition_exchange_archive         [ disabled ]  MDEV-5077 2013-09-27 svoj
 
 Cannot exchange partition with archive table
 
main.1st                                 w5 [ pass ]      1
 
connect.alter_xml                        w2 [ skipped ]  Need windows
 
connect.alter_xml2                       w3 [ pass ]     20
 
main.gis                                 w1 [ pass ]    511

Or you can cherry-pick commit 6fab256bc85 (that I, unfortunately, have completely forgotten of) into 10.3

Comment by Nayuta Yanagisawa (Inactive) [ 2021-07-26 ]

serg Thank you for your review. I will cherry-pick 6fab256bc85.

Comment by Nayuta Yanagisawa (Inactive) [ 2021-07-26 ]

You say "to 10.3" which I understand to be a mistake for "to 10.4".

Comment by Nayuta Yanagisawa (Inactive) [ 2021-07-26 ]

Closed by https://github.com/MariaDB/server/commit/389f5cf76fa2bd5e3dacd074395db4d0ef59c9f2.

Generated at Thu Feb 08 09:12:03 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.