Code: Select all
$ systemd-analyze blame 7.651s dhcpcd.service 1.647s dev-mmcblk0p2.device 1.403s apt-daily-upgrade.service 1.383s apt-daily.service 716ms dphys-swapfile.service 633ms keyboard-setup.service 578ms systemd-fsck-root.service 543ms raspi-config.service 492ms systemd-logind.service 444ms networking.service
This seems possible to me. If all 128 Pi computers hit the same NFS server at the same time when shutting down, then the server better have been configured for that kind of load. What may happen is that the NFS server subsequently appears offline for some of the Pi's which then wait until the server recovers so they can unmount their network mounts.