#1053485 mariadb: FTBFS on sparc64: Post-build test suite fails on main.ctype_binary main.ctype_latin1

Package:
src:mariadb
Source:
src:mariadb
Submitter:
Otto Kekäläinen
Date:
2025-12-10 07:07:03 UTC
Severity:
normal
Tags:
#1053485#5
Date:
2023-10-05 05:33:35 UTC
From:
To:
Builds on sparc64 repeatedly failed on tests main.ctype_binary
main.ctype_latin1 with error message:

main.ctype_binary                        w13 [ fail ]
        Test ended at 2023-10-04 07:23:33
CURRENT_TEST: main.ctype_binary
mysqltest: At line 209: query 'SELECT COUNT(*) FROM t WHERE
EXTRACTVALUE(c,'a')='a'' failed: <Unknown> (2013): Lost connection to
server during query
..
Thread 1 (Thread 0xffff8001022f68c0 (LWP 508035)):
#0  0xffff8001026918f8 in ?? () from /lib/sparc64-linux-gnu/libc.so.6
No symbol table info available.
#1  0x0000010000a18920 in handle_fatal_signal (sig=10) at
./sql/signal_handler.cc:360
        curr_time = 1696404205
        tm = {tm_sec = 25, tm_min = 23, tm_hour = 7, tm_mday = 4,
tm_mon = 9, tm_year = 123, tm_wday = 3, tm_yday = 276, tm_isdst = 0,
tm_gmtoff = 0, tm_zone = 0x100022e69e0 "UTC"}
        thd = 0xffff80011c000c68
        print_invalid_query_pointer = false
Backtrace stopped: Cannot access memory at address 0xf0

This indicates that the code path in these two tests are able to crash
the server.

Full logs with full stack trace at:
* https://buildd.debian.org/status/fetch.php?pkg=mariadb&arch=sparc64&ver=1%3A10.11.5-1&stamp=1696405551&raw=0
* https://buildd.debian.org/status/fetch.php?pkg=mariadb&arch=sparc64&ver=1%3A10.11.5-1&stamp=1696474965&raw=0

These two tests were passing in the 10.11.4-1 builds in Debian
unstable, so this is clearly a regression in 10.11.5.
#1053485#10
Date:
2023-10-09 04:49:44 UTC
From:
To:
I disabled these tests in
https://salsa.debian.org/mariadb-team/mariadb-server/-/commit/f3b99c24d2577b35530e6106cb527cc94cb1e061
and thus these failures are no longer visible in the 1:10.11.5-2
upload.

However several other tests now crashed: main.partition_order
main.func_str main.func_like main.func_math main.group_by
main.group_by_null main.features main.type_datetime main.xml

See detailed stack traces in
https://buildd.debian.org/status/fetch.php?pkg=mariadb&arch=sparc64&ver=1%3A10.11.5-2&stamp=1696744591&raw=0
#1053485#15
Date:
2023-10-19 04:34:42 UTC
From:
To:
Note that this issue still exists, but is not visible and build is not
failing on it simply due to issue being skipped via the
unstable-tests.* in
https://salsa.debian.org/mariadb-team/mariadb-server/-/blob/debian/latest/debian/
#1053485#25
Date:
2025-04-19 20:29:59 UTC
From:
To:
#1053485#30
Date:
2025-04-19 20:48:04 UTC
From:
To:
#1053485#37
Date:
2025-12-10 07:05:55 UTC
From:
To:
I ran a detailed stack trace on stadler.debian.net and posted it in
upstream Jira at https://jira.mariadb.org/browse/MDEV-36652 in hopes
upstream could gain valuable insights from reading it. I also posted
there my notes on the exact steps I ran on stadler to build MariaDB
and run mariadb-test-run in case someone wants to repeat it. I can
also post the same thing here if you think it is valuable for Debian
bug tracking.