O comando systemd-analyze critical-chain destaca algumas entradas com uma cor vermelha, o que essa cor indica?

Question

A saída do systemd-analysis critical-chain:

multi-user.target @30.366s
 └─docker.service @5.230s +851ms
  └─network-online.target @5.227s
   └─network.target @5.222s
    └─network.service @4.872s +349ms
     └─NetworkManager-wait-online.service @1.438s +3.431s
      └─NetworkManager.service @1.402s +35ms
       └─netcf-transaction.service @1.320s +56ms
        └─basic.target @1.285s
         └─sockets.target @1.285s
          └─virtlockd.socket @1.285s
           └─sysinit.target @1.279s
            └─systemd-update-utmp.service @1.262s +16ms
             └─auditd.service @1.120s +140ms
              └─systemd-tmpfiles-setup.service @1.071s +44ms
               └─rhel-import-state.service @1.016s +52ms
                └─local-fs.target @1.011s
                 └─var-lib-docker-plugins.mount @5.788s
                  └─dev-mapper-vg1\x2droot.device @372ms +260ms

cores certas entradas vermelhas; neste caso, eles são:

docker.service @5.230s +851ms
network.service @4.872s +349ms
NetworkManager-wait-online.service @1.438s +3.431s
NetworkManager.service @1.402s +35ms
netcf-transaction.service @1.320s +56ms
systemd-update-utmp.service @1.262s +16ms
auditd.service @1.120s +140ms
systemd-tmpfiles-setup.service @1.071s +44ms
rhel-import-state.service @1.016s +52ms
dev-mapper-vg1\x2droot.device @372ms +260ms

O que a cor indica?

    

Answer 1

O vermelho apenas ativa entradas que possuem um valor +?ms após o @... . De acordo com a página de manual systemd-analyze :

   systemd-analyze critical-chain [UNIT...]  prints a tree of the
   time-critical chain of units (for each of the specified UNITs or for the
   default target otherwise). The time after the unit is active or started
   is printed after the "@" character. The time the unit takes to start is
   printed after the "+" character. Note that the output might be misleading
   as the initialization of one service might depend on socket activation
   and because of the parallel execution of units.

Por isso, apenas destaca as unidades que realmente demoram a começar, como serviços reais (Docker, ...) como opostas a alvos simples que acabam de ser cumpridos e que não "iniciam".