[LU-18964] performance-sanity test_2: ERROR: open64("file.mdtest.0.0", 66, 0664) failed

• Type: Bug
• Resolution: Fixed
• Priority: Minor
• Fix Version/s: Lustre 2.17.0
• Severity: 3
• Assignee: Fred Dilger
• Reporter: Maloo

Description

This issue was created by maloo for Frederick Dilger <[e-mail]>

This issue relates to the following test suite run: https://testing.whamcloud.com/test_sets/51c8b27a-f8cb-44a1-a6d9-0f1d091f7d0f

test_2 failed with the following error:

test_2 failed with 1

Test session details:
clients: https://build.whamcloud.com/job/lustre-reviews/112764 - 4.18.0-553.46.1.el8_10.x86_64
servers: https://build.whamcloud.com/job/lustre-reviews/112764 - 4.18.0-553.46.1.el8_lustre.x86_64

<<Please provide additional information about the failure here>>

Large files creation performance test
set file size to 1073741824
install: cannot change permissions of '/mnt/lustre/mdtest': No such file or directory
stat: cannot read file system information for '/mnt/lustre/mdtest': No such file or directory
CMD: onyx-117vm9 params=\$(/usr/sbin/lctl get_param mdt.*.enable_remote_dir_gid);
[[ -z \"lustre-MDT0000\" ]] && param= ||
param=\$(grep lustre-MDT0000 <<< \"\$params\");
[[ -z \$param ]] && param=\"\$params\";
while read s; do echo mds1 \$s;
done <<< \"\$param\"
CMD: onyx-117vm10 params=\$(/usr/sbin/lctl get_param mdt.*.enable_remote_dir_gid);
[[ -z \"lustre-MDT0001\" ]] && param= ||
param=\$(grep lustre-MDT0001 <<< \"\$params\");
[[ -z \$param ]] && param=\"\$params\";
while read s; do echo mds2 \$s;
done <<< \"\$param\"
CMD: onyx-117vm9 params=\$(/usr/sbin/lctl get_param mdt.*.enable_remote_dir_gid);
[[ -z \"lustre-MDT0002\" ]] && param= ||
param=\$(grep lustre-MDT0002 <<< \"\$params\");
[[ -z \$param ]] && param=\"\$params\";
while read s; do echo mds3 \$s;
done <<< \"\$param\"
CMD: onyx-117vm10 params=\$(/usr/sbin/lctl get_param mdt.*.enable_remote_dir_gid);
[[ -z \"lustre-MDT0003\" ]] && param= ||
param=\$(grep lustre-MDT0003 <<< \"\$params\");
[[ -z \$param ]] && param=\"\$params\";
while read s; do echo mds4 \$s;
done <<< \"\$param\"
CMD: onyx-117vm10,onyx-117vm9 /usr/sbin/lctl set_param mdt.*.enable_remote_dir_gid=-1
mdt.lustre-MDT0000.enable_remote_dir_gid=-1
mdt.lustre-MDT0002.enable_remote_dir_gid=-1
mdt.lustre-MDT0001.enable_remote_dir_gid=-1
mdt.lustre-MDT0003.enable_remote_dir_gid=-1
+ chmod 0777 /mnt/lustre
drwxrwxrwx 3 root root 4096 Apr 24 17:34 /mnt/lustre
+ su mpiuser bash -c "/usr/lib64/openmpi/bin/mpirun --mca btl tcp,self --mca btl_tcp_if_include eth0 -mca boot ssh --oversubscribe -machinefile /tmp/auster.machines -np 1 -npernode 2 /usr/lib64/openmpi/bin/mdtest -w=1073741824 -d=/mnt/lustre/mdtest -n=16 -F -R "
onyx-139vm1.onyx.whamcloud.com:rank0.mdtest: Failed to get eth0 (unit 0) cpu set
onyx-139vm1.onyx.whamcloud.com:rank0: PSM3 can't open nic unit: 0 (err=23)
--------------------------------------------------------------------------
Open MPI failed an OFI Libfabric library call (fi_endpoint). This is highly
unusual; your job may behave unpredictably (and/or abort) after this.
Local host: onyx-139vm1
Location: mtl_ofi_component.c:513
Error: Invalid argument (22)
--------------------------------------------------------------------------
– started at 04/24/2025 17:34:21 –
mdtest-4.0.0 was launched with 1 total task(s) on 1 node(s)
Command line used: /usr/lib64/openmpi/bin/mdtest '-w=1073741824' '-d=/mnt/lustre/mdtest' '-n=16' '-F' '-R'
WARNING: Unable to create test directory path /mnt/lustre/mdtest
Path : /mnt/lustre/mdtest
FS : 58.7 GiB Used FS: 0.0% Inodes: 3.0 Mi Used Inodes: 0.0%
Nodemap: 1
random seed: 1745516061
1 tasks, 16 files
WARNING: Unable to create test directory /mnt/lustre/mdtest/test-dir.0-0
WARNING: unable to create tree directory '/mnt/lustre/mdtest/test-dir.0-0/mdtest_tree.0/'
ERROR: open64("/mnt/lustre/mdtest/test-dir.0-0/mdtest_tree.0/file.mdtest.0.0", 66, 0664) failed. Error: No such file or directory, (aiori-POSIX.c:570)

VVVVVVV DO NOT REMOVE LINES BELOW, Added by Maloo for auto-association VVVVVVV
performance-sanity test_2 - test_2 failed with 1