[RESOLVED] long shutdown time

[rychok]

Recently my HDD died and I decided to replace on my desktop well working Mageia 3 with Mageia 5.
Installation went fine and everything was working correctly, but at some moment of updates and adding programs. ( This computer has installed few servers and lot of software sometimes just too check them) the shutdown is taking now about 5 minutes. I had same problem for a while on one of my laptops (Asus GL752, but since there are much more problems with that gaming laptop I have lived with that) but recently is back to normal.

few last lines from output of command journalctl -ab-1 shows that everything is fine till stopping name resolution and after that computer just sits for few minutes , fan is a little loud so it is doing something, but there is no message what actually.. The last message on screen shown for few minutes also confirms stopping name resolution service.

Sep 28 19:01:32 amd64FX.rknet.net ifplugd(enp9s0)[1525]: Exiting.
Sep 28 19:01:32 amd64FX.rknet.net network[12900]: Shutting down interface enp9s0: [ OK ]
Sep 28 19:01:33 amd64FX.rknet.net network[12900]: Shutting down loopback interface: [ OK ]
Sep 28 19:01:33 amd64FX.rknet.net network[12900]: net.ipv4.tcp_syncookies = 0
Sep 28 19:01:33 amd64FX.rknet.net resolvconf[13276]: Stopping resolvconf: [ OK ]
Sep 28 19:06:17 amd64FX.rknet.net umount[13307]: umount: /var: target is busy
Sep 28 19:06:17 amd64FX.rknet.net umount[13307]: (In some cases useful info about processes that
Sep 28 19:06:17 amd64FX.rknet.net umount[13307]: use the device is found by lsof(8) or fuser(1).)
Sep 28 19:06:17 amd64FX.rknet.net systemd[1]: var.mount mount process exited, code=exited status=32
Sep 28 19:06:17 amd64FX.rknet.net systemd[1]: Failed unmounting /var.
Sep 28 19:06:17 amd64FX.rknet.net systemd[1]: Shutting down.
Sep 28 19:06:17 amd64FX.rknet.net systemd-journal[516]: Journal stopped

I'm running kernel 4.1.15 since the 4.4 has no Nvidia driver.
Any help would be appreciated.

[doktor5000]

You should probably at least post the complete output of

Code: Select all

systemctl --no-pager

for a fresh boot, and also attach complete

Code: Select all

journalctl -ab-1

output when you do a fresh boot, wait a minute or so (and mention at what timestamp the boot was completed and you waited) and then simply do a shutdown.

Seems like /var cannot be unmounted as something is still accessing it, my gues would be either bind/named/dnsmasq or something else that uses /var ... do you have a caching DNS server set up, or do you require one?

[rychok]

I looked into rc5.d to check order for shutdown , and found out that next is vmware , and vmware-workstation-server is running. I couldn't stop it but I have disable it for next boot, and shotdown is back to few seconds.