keyctl05 from ubuntu_ltp_syscalls failed on J-oem-6.1

Bug #2055536 reported by Po-Hsu Lin
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
ubuntu-kernel-tests
New
Undecided
Unassigned

Bug Description

Issue found on J-oem-6.1.0-1035-oem

INFO: Test start time: Thu Feb 29 22:33:22 UTC 2024
COMMAND: /opt/ltp/bin/ltp-pan -q -e -S -a 125203 -n 125203 -f /tmp/ltp-sz5PA5vhqa/alltests -l /dev/null -C /dev/null -T /dev/null
LOG File: /dev/null
FAILED COMMAND File: /dev/null
TCONF COMMAND File: /dev/null
Running tests.......
tst_test.c:1690: TINFO: LTP version: 20230929
tst_test.c:1574: TINFO: Timeout per run is 0h 00m 30s
tst_fips.c:22: TINFO: FIPS: off
keyctl05.c:131: TINFO: Try to update the 'asymmetric' key...
keyctl05.c:143: TPASS: updating 'asymmetric' key expectedly failed with EOPNOTSUPP
keyctl05.c:114: TFAIL: unexpected error adding 'dns_resolver' key: EINVAL (22)
keyctl05.c:179: TINFO: Try to update the 'user' key...
keyctl05.c:188: TPASS: didn't crash while racing to update 'user' key

HINT: You _MAY_ be missing kernel fixes:

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=63a0b0509e70

Summary:
passed 2
failed 1
broken 0
skipped 0
warnings 0
INFO: ltp-pan reported some tests FAIL
 on openstack.

Test log:
INFO: Test start time: Thu Feb 29 22:33:22 UTC 2024
COMMAND: /opt/ltp/bin/ltp-pan -q -e -S -a 125203 -n 125203 -f /tmp/ltp-sz5PA5vhqa/alltests -l /dev/null -C /dev/null -T /dev/null
LOG File: /dev/null
FAILED COMMAND File: /dev/null
TCONF COMMAND File: /dev/null
Running tests.......
tst_test.c:1690: TINFO: LTP version: 20230929
tst_test.c:1574: TINFO: Timeout per run is 0h 00m 30s
tst_fips.c:22: TINFO: FIPS: off
keyctl05.c:131: TINFO: Try to update the 'asymmetric' key...
keyctl05.c:143: TPASS: updating 'asymmetric' key expectedly failed with EOPNOTSUPP
keyctl05.c:114: TFAIL: unexpected error adding 'dns_resolver' key: EINVAL (22)
keyctl05.c:179: TINFO: Try to update the 'user' key...
keyctl05.c:188: TPASS: didn't crash while racing to update 'user' key

HINT: You _MAY_ be missing kernel fixes:

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=63a0b0509e70

Summary:
passed 2
failed 1
broken 0
skipped 0
warnings 0
INFO: ltp-pan reported some tests FAIL

Take a look back into previous cycles we can see that this test was not properly tested after -1028:
  * s20240108: 6.1.0-1027.27 (Not tested)
  * 20240108: 6.1.0-1032.32 (Not tested)
  * s20231030: 6.1.0-1029.29 (Not tested)
  * 20231030: 6.1.0-1028.28 (passed)
  * 20231030: 6.1.0-1027.27 (passed)

It should be tested with -1032 since it's a regular cycle. But it looks like tests were not fully generated at that time.

Manual test shows:
  * 6.1.0-1034-oem FAILED
  * 6.1.0-1033-oem FAILED
  * 6.1.0-1032-oem FAILED
  * 6.1.0-1029-oem PASSED

Therefore this is not a regression with 6.1.0-1035-oem this cycle.

Po-Hsu Lin (cypressyew)
tags: added: sru-20240205
Revision history for this message
Po-Hsu Lin (cypressyew) wrote :

This issue can be found on Jammy 5.15.0-102.112 as well.

It can be fixed by commit 759a0e1195 ("keyctl05: increase dns_res_payload data due to kernel code changes") upstream.

Po-Hsu Lin (cypressyew)
tags: added: ltp-20240312
Revision history for this message
Po-Hsu Lin (cypressyew) wrote :

This is failing on Mantic, as well, both the old and the update LTP test suite will fail with the -proposed kernel 6.5.0-27.28 (both passed with 6.5.0-25.25)

According to commit 4e94000 ("keyctl05: Improve the dns_res_payload for boundary testing") in LTP [1], it mentioned that:

> kernel commit acc657692aed ("keys, dns: Fix size check of V1 server-list
> header") from upcoming v6.8-rc1 (going to be backported to stable),
> which fixes the add_key syscall incoming data boundary size issue,

I can see this commit in our Mantic tree but not tagged with any version. We can expect this should be fixed in the next iteration.

[1] https://github.com/linux-test-project/ltp/commit/4e9400050683051c898db674bf0ba8d8ca9d6d2c

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.