Skip to content
Snippets Groups Projects
Commit b623fb8c authored by Scott Jackson's avatar Scott Jackson Committed by Albert Gil
Browse files

Testsuite - Improve temporary debug for test3.1 increasing logging

This is removing some of the temporary debug code added on 3e02d9e6.

Bug 9647
parent 6f01d264
No related branches found
No related tags found
No related merge requests found
...@@ -32,20 +32,18 @@ set file_in "test$test_id.input" ...@@ -32,20 +32,18 @@ set file_in "test$test_id.input"
set job_id 0 set job_id 0
# TODO: Temporary variables to debug bug 9647 (remove once fixed) # TODO: Temporary variables to debug bug 9647 (remove once fixed)
set username [get_my_user_name] set config_dir [get_conf_path]
set hostname [string trimright [run_command_output "hostname -s"]]
proc cleanup {} { proc cleanup {} {
global job_id bin_rm file_in global job_id bin_rm file_in
# TODO: Temporary cleanup to debug bug 9647 (remove once fixed)
global hostname username
if {$hostname eq "knl2" && $username eq "jenkins"} {
run_command "sudo systemctl stop auditd"
}
cancel_job $job_id cancel_job $job_id
exec $bin_rm -f $file_in exec $bin_rm -f $file_in
# TODO: Temporary cleanup of debug for bug 9647 (remove once fixed)
global config_dir scontrol
restore_conf $config_dir/slurm.conf
run_command "$scontrol reconfigure"
} }
# #
...@@ -56,10 +54,11 @@ if {[string compare [get_admin_level] "Administrator"]} { ...@@ -56,10 +54,11 @@ if {[string compare [get_admin_level] "Administrator"]} {
skip "This test can be run only for Administrators" skip "This test can be run only for Administrators"
} }
# TODO: Temporarily start auditing to debug bug 9647 (remove once fixed) # TODO: Temporarily increase logging to debug bug 9647 (remove once fixed)
if {$hostname eq "knl2" && $username eq "jenkins"} { save_conf $config_dir/slurm.conf
run_command "sudo systemctl start auditd" run_command -none "$bin_echo SlurmctldDebug=debug3 >> $config_dir/slurm.conf"
} run_command -none "$bin_echo SlurmdDebug=debug3 >> $config_dir/slurm.conf"
# Allow the test's existing reconfigure call to establish these values
# #
# Have a job running when the reconfigure happens # Have a job running when the reconfigure happens
...@@ -89,23 +88,11 @@ if {[wait_for_job $job_id RUNNING] != 0} { ...@@ -89,23 +88,11 @@ if {[wait_for_job $job_id RUNNING] != 0} {
fail "Waiting for job $job_id to start" fail "Waiting for job $job_id to start"
} }
# TODO: Temporary command to debug bug 9647 (remove once fixed)
if {$hostname eq "knl2" && $username eq "jenkins"} {
run_command_output "$scontrol listpids $job_id node1"
}
# #
# Issue scontrol reconfigure # Issue scontrol reconfigure
# #
reconfigure -fail reconfigure -fail
if {[wait_for_job $job_id RUNNING] != 0} { if {[wait_for_job $job_id RUNNING] != 0} {
# TODO: Temporary audit logging to debug bug 9647 (remove once fixed)
if {$hostname eq "knl2" && $username eq "jenkins"} {
set jobs_dict [get_jobs $job_id]
if {[dict get $jobs_dict $job_id "JobState"] eq "FAILED" && [dict get $jobs_dict $job_id "ExitCode"] eq "137:0"} {
run_command "sudo ausearch -sc kill -k audit_kill -i -ts recent -te now"
}
}
fail "Job $job_id not running after reconfiguration" fail "Job $job_id not running after reconfiguration"
} }
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment