mpoquet
commented
7 years ago Cannot reproduce the issue.
Your scheduler does not seem to handle homogeneous MSG jobs, the provided workload is probably wrong.
Yaml to reproduce:
base_output_directory: /tmp/batsim_tests/issue36
base_variables:
batsim_dir: ${base_working_directory}
implicit_instances:
implicit:
sweep:
platform :
- {"name":"cluster", "filename":"${batsim_dir}/platforms/cluster_issue36.xml", "master_host":"master_host0"}
workload :
- {"name":"issue36", "filename": "${batsim_dir}/workload_profiles/issue36.json"}
#- {"name":"medium", "filename": "${batsim_dir}/workload_profiles/batsim_paper_workload_example.json"}
#- {"name":"tiny", "filename": "${batsim_dir}/workload_profiles/one_delay_job.json"}
algo:
- {"name":"nodegrp", "algo_name":"nodegrp"}
generic_instance:
timeout: 60
working_directory: ${base_working_directory}
output_directory: ${base_output_directory}/results/${algo[name]}_${workload[name]}_${platform[name]}
batsim_command: batsim -p ${platform[filename]} -w ${workload[filename]} -e ${output_directory}/out --config ${output_directory}/batsim.conf -m ${platform[master_host]}
sched_command: cd /home/carni/proj/rust/schedulers && RUST_BACKTRACE=1 cargo run --bin ${algo[algo_name]}
commands_before_execution:
# Generate Batsim config file
- |
#!/usr/bin/env bash
cat > ${output_directory}/batsim.conf << EOF
{
"job_submission": {
"forward_profiles": true,
"from_scheduler":{
"enabled": true,
"acknowledge": true
}
}
}
EOF
commands_before_instances:
- ${batsim_dir}/test/is_batsim_dir.py ${base_working_directory}
- ${batsim_dir}/test/clean_output_dir.py ${base_output_directory}Output:
2017-06-07 18:48:34,176 INFO: Variables = {'algo': {'algo_name': 'nodegrp', 'name': 'nodegrp'}, 'base_output_directory': '/tmp/batsim_tests/issue36', 'base_working_directory': '/home/carni/proj/batsim', 'batsim_dir': '${base_working_directory}', 'instance_id': '402b9074', 'instance_number': 0, 'platform': {'filename': '${batsim_dir}/platforms/cluster_issue36.xml', 'master_host': 'master_host0', 'name': 'cluster'}, 'workload': {'filename': '${batsim_dir}/workload_profiles/issue36.json', 'name': 'issue36'}, 'working_directory': '/home/carni/proj/batsim', 'output_directory': '/tmp/batsim_tests/issue36/results/nodegrp_issue36_cluster'}
2017-06-07 18:48:34,176 INFO: Working directory: /home/carni/proj/batsim
2017-06-07 18:48:34,176 INFO: Output directory: /tmp/batsim_tests/issue36/results/nodegrp_issue36_cluster
2017-06-07 18:48:34,176 INFO: Executing command 'command0'
2017-06-07 18:48:34,184 INFO: command0 finished
2017-06-07 18:48:34,186 INFO: Batsim command: "batsim -p ${platform[filename]} -w ${workload[filename]} -e ${output_directory}/out --config ${output_directory}/batsim.conf -m ${platform[master_host]}"
2017-06-07 18:48:34,186 INFO: Sched command: "cd /home/carni/proj/rust/schedulers && RUST_BACKTRACE=1 cargo run --bin ${algo[algo_name]}"
2017-06-07 18:48:34,186 INFO: Waiting for socket 'tcp://localhost:28000' to be usable
2017-06-07 18:48:34,193 INFO: Socket tcp://localhost:28000 is now usable
2017-06-07 18:48:34,194 INFO: Running Batsim and Sched
2017-06-07 18:48:35,721 ERROR: Sched finished (returncode=101)
2017-06-07 18:48:35,727 ERROR: Sched stderr:
Finished dev [unoptimized + debuginfo] target(s) in 0.0 secs
Running `target/debug/nodegrp`
thread 'main' panicked at 'ErrorImpl { code: Message("missing field `delay`"), line: 1, column: 260 } full str: {"now":0.002400,"events":[{"timestamp":0.001800,"type":"JOB_SUBMITTED","data":{"job_id":"038390!35383","job":{"profile":"8","res":2,"id":"038390!35383","subtime":0,"walltime":1800.000000},"profile":{"com":0.000000,"type":"msg_par_hg","cpu":800000000.000000}}}]}', /home/carni/proj/rust/rustbatsim/src/libbatsim/batsim.rs:430
stack backtrace:
0: std::sys::imp::backtrace::tracing::imp::unwind_backtrace
1: std::sys_common::backtrace::_print
2: std::panicking::default_hook::{{closure}}
3: std::panicking::default_hook
4: std::panicking::rust_panic_with_hook
5: std::panicking::begin_panic
6: std::panicking::begin_panic_fmt
7: batsim::batsim::read_batsim_message
at /home/carni/proj/rust/rustbatsim/src/libbatsim/batsim.rs:430
8: batsim::batsim::Batsim::get_next_message
at /home/carni/proj/rust/rustbatsim/src/libbatsim/batsim.rs:265
9: batsim::batsim::Batsim::run_simulation
at /home/carni/proj/rust/rustbatsim/src/libbatsim/batsim.rs:297
10: nodegrp::main
at ./nodegroup/src/main.rs:21
11: std::panicking::try::do_call
12: __rust_maybe_catch_panic
13: std::rt::lang_start
14: main
15: __libc_start_main
16: _start
2017-06-07 18:48:35,738 ERROR: Killing remaining processes {1824, 1778, 1780, 1823}
adfaure
To reproduce the bug
You can run my scheduler:
and batsim
The cluster:
And you can use this workload
http://github.com/adfaure/ea2868ce9c152d590573bb778d767b7ehttps://gist.github.com/adfaure/ea2868ce9c152d590573bb778d767b7e