[LU-10094] sanity test_17f: 'ls' fails with "ls: reading directory *: Input/output error"

• Status: Resolved
• Project: Lustre
• Affects Version/s: Lustre 2.11.0, Lustre 2.12.0, Lustre 2.13.0, Lustre 2.10.7, Lustre 2.12.1
• Fix Version/s: Lustre 2.13.0, Lustre 2.12.3
• Type: Bug
• Priority: Critical
• Reporter: James Casper
• Assignee: Lai Siyao
• Resolution: Fixed
• Labels: ppc

Environment

trevis, full, x86_64 servers, ppc clients
servers: el7.4, ldiskfs, branch master, v2.10.53.1, b3642
clients: el7.4, branch master, v2.10.53.1, b3642

• Severity: 3

Description

https://testing.whamcloud.com/test_sessions/ba995751-659c-4e63-9b5b-fbf101137b78

From test_log:

ls: reading directory /mnt/lustre/d17f.sanity: Input/output error
 sanity test_17f: @@@@@@ FAIL: test_17f failed with 2 
  Trace dump:
  = /usr/lib64/lustre/tests/test-framework.sh:5289:error()
  = /usr/lib64/lustre/tests/test-framework.sh:5565:run_one()
  = /usr/lib64/lustre/tests/test-framework.sh:5604:run_one_logged()
  = /usr/lib64/lustre/tests/test-framework.sh:5451:run_test()
  = /usr/lib64/lustre/tests/sanity.sh:459:main()

Issue Links

• is related to: LU-10095 sanity test_18: Failed to ls /mnt/lus...
• is related to: LU-10096 sanity test_22: ls -lR /mnt/lustre/d2...
• is related to: LU-10983 sanity tests fails with 'ls: reading ...
• is related to: LU-10100 sanity test_27a: setstripe failed wit...
• is related to: LU-10099 sanity test_24A: Expected 5000 files,...

Comments

[James Nunez (Inactive)]

Several sanity tests have 'ls' fail with the "reading directory ... Input/output error" for PPC architectures including test_17f, 18, 22, 24v, 24A, 32b, 32d, 32f, 32h, 48a, 48b, 48c, 51a, 51b, 56ab, and 154a.

For full test group results, the first time we see these tests fail for PPC is on 2017-09-17 20:43:36 UTC for master build # 3642, version 2.10.53.1.

[Jian Yu]

On ppc64 client:

# ls /mnt/lustre/
ls: reading directory /mnt/lustre/: Input/output error

Dmesg:

[53353.003090] Lustre: Mounted lustre-client
[53354.406948] Lustre: DEBUG MARKER: Using TIMEOUT=20
[53372.104367] Lustre: 30604:0:(mdc_request.c:1549:mdc_read_page()) Page-wide hash collision: 0xfeffffffffffffff
[53378.035937] Lustre: lustre-OST0000-osc-c0000000788f6800: disconnect after 24s idle
[54485.730632] Lustre: 30675:0:(mdc_request.c:1549:mdc_read_page()) Page-wide hash collision: 0xfeffffffffffffff
[54485.730769] Lustre: 30675:0:(mdc_request.c:1549:mdc_read_page()) Skipped 1 previous similar message

[Lai Siyao]

Can you run 'getconf PAGE_SIZE' on ppc64 client?

[Gerrit Updater]

Lai Siyao ([e-mail]) uploaded a new patch: https://review.whamcloud.com/35517
Subject: LU-10094 mdc: dir page ldp_hash_end mistakenly adjusted
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 5fcc4cc687eb995946fad3bdf4e96caf4264b20c

[Jian Yu]

Can you run 'getconf PAGE_SIZE' on ppc64 client?

# uname -m
ppc64
# getconf PAGE_SIZE
65536

[Lai Siyao]

Mmm, the above patch should be able to fix this issue.

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch https://review.whamcloud.com/35517/
Subject: LU-10094 mdc: dir page ldp_hash_end mistakenly adjusted
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: d8b19ae6617733df003a906aca1791791a5f0eff

[Peter Jones]

Landed for 2.13

[Gerrit Updater]

Jian Yu ([e-mail]) uploaded a new patch: https://review.whamcloud.com/35812
Subject: LU-10094 mdc: dir page ldp_hash_end mistakenly adjusted
Project: fs/lustre-release
Branch: b2_12
Current Patch Set: 1
Commit: 3511039df9f17f959fc57c4046ad00984824886d

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch https://review.whamcloud.com/35812/
Subject: LU-10094 mdc: dir page ldp_hash_end mistakenly adjusted
Project: fs/lustre-release
Branch: b2_12
Current Patch Set:
Commit: c7538cf4bf952fd5222e71541fbab544e35a5b77