zram01 / zram02 / zram03 in kernel_misc from ubuntu_ltp failed on D-Oracle

[Po-Hsu Lin]

It looks like the failures in zram01 are causing zram02 and zram03 to fail as well.

Seems that this is broken since commit 7cfac804 (zram: Rewrite shell tests into new API)

 tag=zram01 stime=1573887406 dur=300 exit=signaled stat=9 core=no cu=23757 cs=7590
 startup='Sat Nov 16 06:56:46 2019'
 zram01 1 TINFO: timeout per run is 0h 5m 0s
 zram01 1 TINFO: create '4' zram device(s)
 zram01 1 TPASS: test succeeded
 zram01 1 TCONF: The device attribute max_comp_streams was introduced in kernel 3.15 and deprecated in 4.7
 zram01 2 TINFO: test that we can set compression algorithm
 zram01 2 TINFO: supported algs: [lzo] lz4 lz4hc 842 zstd
 zram01 2 TINFO: /sys/block/zram0/comp_algorithm = 'lzo' (1/4)
 zram01 2 TINFO: /sys/block/zram1/comp_algorithm = 'lzo' (2/4)
 zram01 2 TINFO: /sys/block/zram2/comp_algorithm = 'lzo' (3/4)
 zram01 2 TINFO: /sys/block/zram3/comp_algorithm = 'lzo' (4/4)
 zram01 2 TPASS: test succeeded
 zram01 3 TINFO: set disk size to zram device(s)
 zram01 3 TINFO: /sys/block/zram0/disksize = '26214400' (1/4)
 zram01 3 TINFO: /sys/block/zram1/disksize = '26214400' (2/4)
 zram01 3 TINFO: /sys/block/zram2/disksize = '26214400' (3/4)
 zram01 3 TINFO: /sys/block/zram3/disksize = '402653184' (4/4)
 zram01 3 TPASS: test succeeded
 zram01 4 TINFO: set memory limit to zram device(s)
 zram01 4 TINFO: /sys/block/zram0/mem_limit = '25M' (1/4)
 zram01 4 TINFO: /sys/block/zram1/mem_limit = '25M' (2/4)
 zram01 4 TINFO: /sys/block/zram2/mem_limit = '25M' (3/4)
 zram01 4 TINFO: /sys/block/zram3/mem_limit = '384M' (4/4)
 zram01 4 TPASS: test succeeded
 zram01 5 TINFO: make ext3 filesystem on /dev/zram0
 zram01 5 TINFO: make ext4 filesystem on /dev/zram1
 zram01 5 TINFO: make xfs filesystem on /dev/zram2
 zram01 5 TINFO: make btrfs filesystem on /dev/zram3
 zram01 5 TPASS: zram_makefs succeeded
 zram01 6 TINFO: mount /dev/zram0
 zram01 6 TINFO: mount /dev/zram1
 zram01 6 TINFO: mount /dev/zram2
 zram01 6 TINFO: mount /dev/zram3
 zram01 6 TPASS: mount of zram device(s) succeeded
 zram01 7 TINFO: filling zram0 (it can take long time)
 zram01 7 TPASS: zram0 was filled with '20628' KB
 zram01 7 TPASS: compression ratio: 732.90:1
 zram01 7 TINFO: filling zram1 (it can take long time)
 zram01 7 TPASS: zram1 was filled with '20140' KB
 zram01 7 TPASS: compression ratio: 5151.97:1
 zram01 7 TINFO: filling zram2 (it can take long time)
 zram01 7 TPASS: zram2 was filled with '17756' KB
 zram01 7 TPASS: compression ratio: 285.26:1
 zram01 7 TINFO: filling zram3 (it can take long time)
 zram01 1 TBROK: test killed, timeout! If you are running on slow machine, try exporting LTP_TIMEOUT_MUL > 1

 tag=zram02 stime=1573887712 dur=0 exit=exited stat=1 core=no cu=2 cs=0
 startup='Sat Nov 16 07:01:52 2019'
 zram02 1 TINFO: timeout per run is 0h 5m 0s
 zram02 1 TINFO: create '1' zram device(s)
 zram02 1 TFAIL: unexpected num of devices: 4
 /opt/ltp/testcases/bin/zram02.sh: 27: echo: echo: I/O error
 zram02 1 TINFO: AppArmor enabled, this may affect test results
 zram02 1 TINFO: it can be disabled with TST_DISABLE_APPARMOR=1 (requires super/root)
 zram02 1 TINFO: loaded AppArmor profiles: none

 Summary:
 passed 0
 failed 1
 skipped 0
 warnings 0

 tag=zram03 stime=1573887718 dur=0 exit=exited stat=2 core=no cu=0 cs=0
 startup='Sat Nov 16 07:01:58 2019'
 zram03 0 TINFO : create a zram device with 536870912 bytes in size.
 zram03 1 TBROK : safe_file_ops.c:301: Failed to close FILE '/sys/block/zram0/disksize' at zram03.c:87: errno=EBUSY(16): Device or resource busy
 zram03 2 TBROK : safe_file_ops.c:301: Remaining cases broken

ProblemType: Bug
DistroRelease: Ubuntu 19.04
Package: linux-image-5.0.0-1005-oracle 5.0.0-1005.9
ProcVersionSignature: User Name 5.0.0-1005.9-oracle 5.0.21
Uname: Linux 5.0.0-1005-oracle x86_64
ApportVersion: 2.20.10-0ubuntu27.1
Architecture: amd64
Date: Mon Nov 18 07:50:37 2019
SourcePackage: linux-signed-oracle
UpgradeStatus: No upgrade log present (probably fresh install)

[Po-Hsu Lin (cypressyew)]

This issue can be found on AWS Eoan 5.3 as well.

It looks like bumping the timeout to 15 min (LTP_TIMEOUT_MUL=3) can solve this issue.

[Po-Hsu Lin (cypressyew)]

On AWS Eoan, it's not failing across all the instances, just:
3/3 ARM64 instances:
  * a1.2xlarge
  * a1.large
  * a1.medium

6/15 AMD64 instances:
  * c3.xlarge
  * m3.large
  * m4.large
  * r3.large
  * t3a.medium
  * x1e.xlarge

[Po-Hsu Lin (cypressyew)]

Tested on Oracle (Disco) / AWS (Eoan)

LTP_TIMEOUT_MUL=3 works, I will apply the patch for this.

[Po-Hsu Lin (cypressyew)]

https://kernel.ubuntu.com/git/ubuntu/autotest-client-tests.git/commit/?id=135585f1ad75dcea0e5bab344a7c60fffbb15dfb

[Po-Hsu Lin (cypressyew)]

Hi Sean,
I found the ACT repo on AWS is out-dated, if you found this issue on it please give it another try.
Thanks.

[Kelsey Steele (kelsey-steele)]

Spotted on Focal/azure : 5.4.0-1029.29 : amd64

[Kelsey Steele (kelsey-steele)]

failed on Bionic/azure-4.15 : 4.15.0-1097.107 : amd64