Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
S
Slurm
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Build
Pipelines
Jobs
Pipeline schedules
Artifacts
Deploy
Releases
Package registry
Model registry
Operate
Environments
Terraform modules
Monitor
Incidents
Service Desk
Analyze
Value stream analytics
Contributor analytics
CI/CD analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Terms and privacy
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
tud-zih-energy
Slurm
Commits
71fbb7ea
Commit
71fbb7ea
authored
4 years ago
by
Scott Jackson
Committed by
Albert Gil
4 years ago
Browse files
Options
Downloads
Patches
Plain Diff
Testsuite - Add temporary debug traces to test 19.4 to debug bug 9665
Bug 9665
parent
eb1b903c
No related branches found
No related tags found
No related merge requests found
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
testsuite/expect/test19.4
+44
-61
44 additions, 61 deletions
testsuite/expect/test19.4
with
44 additions
and
61 deletions
testsuite/expect/test19.4
+
44
−
61
View file @
71fbb7ea
...
...
@@ -32,11 +32,9 @@ set exit_code 0
set file_in "test$test_id.input"
set file_out "test$test_id.output"
proc cleanup {} {
global bin_rm file_in file_out
exec $bin_rm -f $file_in $file_out
}
# TODO: Temporary variables to debug bug 9665 (remove once fixed)
set trigger_get_failed false
set config_dir [get_conf_path]
#
# get my uid and clear any vestigial triggers
...
...
@@ -46,8 +44,6 @@ if {$uid == 0} {
skip "This test can't be run as user root"
}
exec $strigger --clear --quiet --user=$uid
#
# Build input script file
#
...
...
@@ -55,52 +51,50 @@ set cwd "[$bin_pwd]"
exec $bin_rm -f $file_in $file_out
make_bash_script $file_in "$bin_echo RECONFIG >$cwd/$file_out"
set disabled 0
set matches 0
set strigger_pid [spawn $strigger --set -v -offset=0 --reconfig --program=$cwd/$file_in]
expect {
-re "permission denied" {
set disabled 1
exp_continue
}
-re "trigger set" {
incr matches
exp_continue
}
timeout {
slow_kill $strigger_pid
fail "strigger not responding"
}
eof {
wait
proc cleanup {} {
global bin_rm file_in file_out strigger uid
# TODO: Temporary debug for bug 9665 (remove once fixed)
global trigger_get_failed
if {$trigger_get_failed} {
if [
wait_for -timeout 5 -poll .1 {
regexp "$file_in" [run_command_output "$strigger --get -v --reconfig --user=$uid"]
} {}
] {
log_warn "Trigger get failed and never recovered"
} else {
log_warn "Trigger get failed but became available afterwards"
}
}
global config_dir scontrol
restore_conf $config_dir/slurm.conf
run_command "$scontrol reconfigure"
run_command -none -nolog "$strigger --clear --reconfig --user=$uid --quiet"
exec $bin_rm -f $file_in $file_out
}
if {$disabled == 1} {
exec $bin_rm -f $file_in
skip "Current configuration prevents setting triggers. Need to run as SlurmUser or make SlurmUser=root"
}
if {$matches == 0} {
# TODO: Temporarily increase logging to debug bug 9665 (remove once fixed)
save_conf $config_dir/slurm.conf
run_command -none "$bin_echo SlurmctldDebug=debug3 >> $config_dir/slurm.conf"
run_command -none "$bin_echo DebugFlags=Triggers >> $config_dir/slurm.conf"
run_command -fail "$scontrol reconfigure"
wait_for -timeout 5 -pollinterval .1 {[get_config_param "SlurmctldDebug"] eq "debug3"} {}
wait_for -timeout 5 -pollinterval .1 {[get_config_param "DebugFlags"] eq "Triggers"} {}
run_command -fail "$strigger --clear --quiet --user=$uid"
# Create a trigger on a reconfig event
if {![regexp "trigger set" [run_command_output -fail "$strigger --set -v --offset=0 --reconfig --program=$cwd/$file_in"]]} {
fail "Trigger creation failure"
}
set matches 0
set strigger_pid [spawn $strigger --get -v --reconfig --user=$uid]
expect {
-re "$file_in" {
incr matches
exp_continue
}
timeout {
slow_kill $strigger_pid
fail "strigger not responding"
}
eof {
wait
}
}
if {$matches == 0} {
log_error "Trigger get failure"
set exit_code 1
# Verify that the trigger has been created
if {![regexp "$file_in" [run_command_output -fail "$strigger --get -v --reconfig --user=$uid"]]} {
# TODO: Temporary variable to debug bug 9665 (remove once fixed)
set trigger_get_failed true
fail "Trigger get failure"
}
#
...
...
@@ -118,10 +112,10 @@ expect {
}
}
if {$invalid == 0} {
#
Add delay for slurmctld to process triggers (every 15 secs)
#
Add delay for slurmctld to process triggers (every 15 secs)
exec sleep 15
#
Plus time for cross-platform NSF delays
#
Plus time for cross-platform NSF delays
if {[wait_for_file $file_out] != 0} {
log_error "File $file_out is missing"
set exit_code 1
...
...
@@ -144,17 +138,6 @@ if {$invalid == 0} {
}
}
set strigger_pid [spawn $strigger --clear -v --reconfig --user=$uid]
expect {
timeout {
slow_kill $strigger_pid
fail "strigger not responding"
}
eof {
wait
}
}
if {$exit_code != 0} {
fail "Test failed due to previous errors (\$exit_code = $exit_code)"
}
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment