uevent:uevent01 in ubuntu_ltp failed on F-gcp n2d-standard-64

[Po-Hsu Lin]

Issue spotted on F-gcp / F-gcp-fips only on GCP instance n2d-standard-64

This is not a regression, issue can be found with 5.4.0-1093.102 in sru-20221010

Test is not actually failing, just issued a warning and thus return 1

 INFO: Test start time: Wed Oct 26 09:25:11 UTC 2022
 COMMAND: /opt/ltp/bin/ltp-pan -q -e -S -a 557151 -n 557151 -f /tmp/ltp-aoiH1noOI0/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:1526: TINFO: Timeout per run is 0h 00m 30s
 tst_device.c:89: TINFO: Found free device 5 '/dev/loop5'
 uevent01.c:24: TINFO: Attaching device /dev/loop5
 uevent.h:49: TINFO: Got uevent:
 uevent.h:52: TINFO: change@/devices/virtual/block/loop5
 uevent.h:52: TINFO: ACTION=change
 uevent.h:52: TINFO: DEVPATH=/devices/virtual/block/loop5
 uevent.h:52: TINFO: SUBSYSTEM=block
 uevent.h:52: TINFO: MAJOR=7
 uevent.h:52: TINFO: MINOR=5
 uevent.h:52: TINFO: DEVNAME=loop5
 uevent.h:52: TINFO: DEVTYPE=disk
 uevent.h:52: TINFO: SEQNUM=4941557
 uevent.h:140: TPASS: Got expected UEVENT
 uevent01.c:26: TINFO: Detaching device /dev/loop5
 uevent.h:49: TINFO: Got uevent:
 uevent.h:52: TINFO: change@/devices/virtual/block/loop5
 uevent.h:52: TINFO: ACTION=change
 uevent.h:52: TINFO: DEVPATH=/devices/virtual/block/loop5
 uevent.h:52: TINFO: SUBSYSTEM=block
 uevent.h:52: TINFO: SYNTH_UUID=0
 uevent.h:52: TINFO: MAJOR=7
 uevent.h:52: TINFO: MINOR=5
 uevent.h:52: TINFO: DEVNAME=loop5
 uevent.h:52: TINFO: DEVTYPE=disk
 uevent.h:52: TINFO: SEQNUM=4941558
 uevent.h:140: TPASS: Got expected UEVENT
 tst_device.c:255: TWARN: ioctl(/dev/loop5, LOOP_CLR_FD, 0) no ENXIO for too long

 Summary:
 passed 2
 failed 0
 broken 0
 skipped 0
 warnings 1

Nothing interesting from dmesg:
[ 1076.444906] LTP: starting uevent01
(nothing afterward)

[Po-Hsu Lin (cypressyew)]

Tested with the latest LTP (548781c1), this issue still exist with 5.4.0-1102-gcp.

Tested with 5.4.0-1090-gcp, this issue still exist:
Running tests.......
<<<test_start>>>
tag=uevent01 stime=1678951840
cmdline="uevent01"
contacts=""
analysis=exit
<<<test_output>>>
incrementing stop
tst_test.c:1558: TINFO: Timeout per run is 0h 00m 30s
tst_device.c:93: TINFO: Found free device 8 '/dev/loop8'
uevent01.c:24: TINFO: Attaching device /dev/loop8
uevent.h:49: TINFO: Got uevent:
uevent.h:52: TINFO: change@/devices/virtual/block/loop8
uevent.h:52: TINFO: ACTION=change
uevent.h:52: TINFO: DEVPATH=/devices/virtual/block/loop8
uevent.h:52: TINFO: SUBSYSTEM=block
uevent.h:52: TINFO: MAJOR=7
uevent.h:52: TINFO: MINOR=8
uevent.h:52: TINFO: DEVNAME=loop8
uevent.h:52: TINFO: DEVTYPE=disk
uevent.h:52: TINFO: SEQNUM=7694
uevent.h:140: TPASS: Got expected UEVENT
uevent01.c:26: TINFO: Detaching device /dev/loop8
uevent.h:49: TINFO: Got uevent:
uevent.h:52: TINFO: change@/devices/virtual/block/loop8
uevent.h:52: TINFO: ACTION=change
uevent.h:52: TINFO: DEVPATH=/devices/virtual/block/loop8
uevent.h:52: TINFO: SUBSYSTEM=block
uevent.h:52: TINFO: SYNTH_UUID=0
uevent.h:52: TINFO: MAJOR=7
uevent.h:52: TINFO: MINOR=8
uevent.h:52: TINFO: DEVNAME=loop8
uevent.h:52: TINFO: DEVTYPE=disk
uevent.h:52: TINFO: SEQNUM=7695
uevent.h:140: TPASS: Got expected UEVENT
tst_device.c:259: TWARN: ioctl(/dev/loop8, LOOP_CLR_FD, 0) no ENXIO for too long

Summary:
passed 2
failed 0
broken 0
skipped 0
warnings 1
$ uname -a
Linux f-lgcp-gcp-5-4-n2dstd64-u-ltp 5.4.0-1090-gcp #98-Ubuntu SMP Mon Oct 3 02:45:32 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux

The version of LTP we’re using for 2023.02.27 is almost the same with f-gcp-5.4.0-1090 in 2022.09.19 (just with one SAUCE patch for fs_fill test from Andrei on top of ac1a3e40c5).

Since this is just a WARNING and we have issue hinted, I will leave this bug open.