Scaphandre does not work on VMs with more than one logical host CPU

[armaniv]

Bug description

Scaphandre does not work on Qemu/KVM virtual machines with more than one logical host CPU assigned.

To Reproduce

1. Create a VM and allocate to it two or more logical host CPUs.
2. Follow the procedure described in the documentation under the Qemu exporter section.
3. Run, for example:
   sudo docker run -v /var/scaphandre:/var/scaphandre -v /proc:/proc -ti hubblo/scaphandre --vm json -t 15

Expected behavior

Scaphandre measures and shows metrics in JSON format.

Environment

• Linux distribution version - Ubuntu 20.04.3 LTS
• Kernel version - 5.11.0-38-generic
• KVM verision - QEMU emulator version 4.2.1 (Debian 1:4.2-3ubuntu6.18)

Additional context

• The command sudo docker run -e RUST_BACKTRACE=1 -v /var/scaphandre:/var/scaphandre -v /proc:/proc -ti hubblo/scaphandre --vm json -t 15 outputs:

thread 'main' panicked at 'Trick: if you are running on a vm, do not forget to use --vm parameter invoking scaphandre at the command line', src/sensors/mod.rs:263:18
stack backtrace:
   0: rust_begin_unwind
             at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/panicking.rs:483
   1: core::panicking::panic_fmt
             at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/core/src/panicking.rs:85
   2: core::option::expect_failed
             at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/core/src/option.rs:1226
   3: scaphandre::sensors::Topology::add_cpu_cores
   4: <scaphandre::sensors::powercap_rapl::PowercapRAPLSensor as scaphandre::sensors::Sensor>::generate_topology
   5: <scaphandre::sensors::powercap_rapl::PowercapRAPLSensor as scaphandre::sensors::Sensor>::get_topology
   6: scaphandre::exporters::json::JSONExporter::runner
   7: scaphandre::run
   8: scaphandre::main
note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace

• The command sudo docker run -e RUST_BACKTRACE=full -v /var/scaphandre:/var/scaphandre -v /proc:/proc -ti hubblo/scaphandre --vm json -t 15 outputs:

thread 'main' panicked at 'Trick: if you are running on a vm, do not forget to use --vm parameter invoking scaphandre at the command line', src/sensors/mod.rs:263:18
stack backtrace:
   0:     0x55a129f5bc70 - std::backtrace_rs::backtrace::libunwind::trace::h72c2fb8038f1bbee
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/../../backtrace/src/backtrace/libunwind.rs:96
   1:     0x55a129f5bc70 - std::backtrace_rs::backtrace::trace_unsynchronized::h1e3b084883f1e78c
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/../../backtrace/src/backtrace/mod.rs:66
   2:     0x55a129f5bc70 - std::sys_common::backtrace::_print_fmt::h3bf6a7ebf7f0394a
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/sys_common/backtrace.rs:79
   3:     0x55a129f5bc70 - <std::sys_common::backtrace::_print::DisplayBacktrace as core::fmt::Display>::fmt::h2e8cb764b7fe02e7
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/sys_common/backtrace.rs:58
   4:     0x55a129f7ec4c - core::fmt::write::h7a1184eaee6a8644
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/core/src/fmt/mod.rs:1080
   5:     0x55a129f55442 - std::io::Write::write_fmt::haeeb374d93a67eac
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/io/mod.rs:1516
   6:     0x55a129f5e11d - std::sys_common::backtrace::_print::h1d14a7f6ad632dc8
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/sys_common/backtrace.rs:61
   7:     0x55a129f5e11d - std::sys_common::backtrace::print::h301abac8bb2e3e81
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/sys_common/backtrace.rs:48
   8:     0x55a129f5e11d - std::panicking::default_hook::{{closure}}::hde0cb80358a6920a
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/panicking.rs:208
   9:     0x55a129f5ddc8 - std::panicking::default_hook::h9b1a691049a0ec8f
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/panicking.rs:227
  10:     0x55a129f5e801 - std::panicking::rust_panic_with_hook::h2bdec87b60580584
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/panicking.rs:577
  11:     0x55a129f5e3a9 - std::panicking::begin_panic_handler::{{closure}}::h101ca09d9df5db47
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/panicking.rs:484
  12:     0x55a129f5c0dc - std::sys_common::backtrace::__rust_end_short_backtrace::h3bb85654c20113ca
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/sys_common/backtrace.rs:153
  13:     0x55a129f5e369 - rust_begin_unwind
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/panicking.rs:483
  14:     0x55a129f7c871 - core::panicking::panic_fmt::h48c31e1e3d550146
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/core/src/panicking.rs:85
  15:     0x55a129f7c603 - core::option::expect_failed::hf3f43f1792267e24
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/core/src/option.rs:1226
  16:     0x55a129a1a897 - scaphandre::sensors::Topology::add_cpu_cores::h63e5518e58e301c8
  17:     0x55a1299b00bd - <scaphandre::sensors::powercap_rapl::PowercapRAPLSensor as scaphandre::sensors::Sensor>::generate_topology::h5bc10914f1385482
  18:     0x55a1299b04fa - <scaphandre::sensors::powercap_rapl::PowercapRAPLSensor as scaphandre::sensors::Sensor>::get_topology::h06a9dd8e7d085121
  19:     0x55a129a2ae9b - scaphandre::exporters::json::JSONExporter::runner::h7f5a768af8fc512d
  20:     0x55a12996b6b9 - scaphandre::run::ha4ed5db2b42f3873
  21:     0x55a1299582f7 - scaphandre::main::hb3cd48c47ec910bd
  22:     0x55a1299567d3 - std::sys_common::backtrace::__rust_begin_short_backtrace::h3c9a2d2e7724640c
  23:     0x55a1299567e9 - std::rt::lang_start::{{closure}}::h829cbaf531010019
  24:     0x55a129f5ed17 - core::ops::function::impls::<impl core::ops::function::FnOnce<A> for &F>::call_once::he179d32a5d10d957
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/core/src/ops/function.rs:259
  25:     0x55a129f5ed17 - std::panicking::try::do_call::hcb3d5e7be089b2b4
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/panicking.rs:381
  26:     0x55a129f5ed17 - std::panicking::try::h7ac93b0cd56fb701
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/panicking.rs:345
  27:     0x55a129f5ed17 - std::panic::catch_unwind::h7b40e396c93a4fcd
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/panic.rs:382
  28:     0x55a129f5ed17 - std::rt::lang_start_internal::h142b9cc66267fea1
                               at /rustc/7eac88abb2e57e752f3302f02be5f3ce3d7adfb4/library/std/src/rt.rs:51
  29:     0x55a1299584a2 - main
  30:     0x7fcd55035cb2 - __libc_start_main
  31:     0x55a12995617a - _start
  32:                0x0 - <unknown>

• The execution of the previous command in a Qemu/KVM virtual machine (ore the same VM resized) with one logical host CPU assigned works as expected.

[bpetit]

Hi, thanks for this bug report. I should work on that in the following weeks.

[gebauerm]

Hi, a colleague and I really appreciate your tool and would like to use it for sustainability research purposes. Unfortunately, we are encountering the same bug, is there any news on this?

[bpetit]

Hi,

In addition to @tawalaya's #133, I'll work on qemu exporter in the same PR as warp10 exporter refactoring : #291

I'll update this thread as well

[bpetit]

should be better in dev now thanks to @tawalaya

could you give it a try ?

[tawalaya]

Great, I'll try it out asap 😃

[tawalaya]

@bpetit tested out the version in dev on our cluster, and nothing broke :) I did not have time to check if the data produced made sense overall the fix looks promising. Thanks 👍