systemd update

Matt · Dec 24, 2017

0014278: systemd update causes /run tmpfs to fill up with sessions - CentOS Bug Tracker

Just troublshooting why my /run directory was full (and getting larger on some other servers).

So it appears we need to treat systemd updates the same as glibc updates, and reboot the server following the update.

pamamolf · Dec 25, 2017

Does the reboot after updating Centos 7 fix that issue and clears all the extra data or we need to clear them manually?

eva2000 · Dec 25, 2017

Interesting issue !
checking one of my OVH CentOS 7.4 servers with Centmin Mod 123.09beta01

Code (Text):

uptime
 20:45:05 up 101 days,  2:14,  1 user,  load average: 0.00, 0.01, 0.05

systemd sessions logged from Oct 19, 2017 to Dec 24, 2017

Code (Text):

ls -lahrt /run/systemd/sessions | head -n3; echo; echo to; echo; ls -lahrt /run/systemd/sessions | tail -3
total 98M
-rw-r--r--  1 root root  263 Oct 19 05:02 14591
-rw-r--r--  1 root root  252 Oct 26 21:25 17840

to

-rw-r--r--  1 root root  299 Dec 24 20:40 42764
-rw-r--r--  1 root root  252 Dec 24 20:45 42765

Code (Text):

du -h --max-depth=1 /run/systemd/
24K     /run/systemd/generator.late
20K     /run/systemd/generator
0       /run/systemd/ask-password-block
98M     /run/systemd/sessions
4.0K    /run/systemd/seats
0       /run/systemd/shutdown
296K    /run/systemd/users
0       /run/systemd/netif
0       /run/systemd/machines
4.0K    /run/systemd/readahead
0       /run/systemd/initctl
4.0K    /run/systemd/journal
0       /run/systemd/ask-password
0       /run/systemd/inaccessible
312K    /run/systemd/system
99M     /run/systemd/

Code (Text):

du -h --max-depth=1 /run
4.0K    /run/amplify-agent
0       /run/runc
8.0K    /run/blkid
4.0K    /run/memcached
12K     /run/docker
4.0K    /run/pcp
4.0K    /run/php-fpm
0       /run/chrony
0       /run/dbus
0       /run/screen
0       /run/svnserve
0       /run/sepermit
0       /run/faillock
0       /run/named
0       /run/ppp
4.0K    /run/tuned
0       /run/netreport
0       /run/net-snmp
0       /run/user
0       /run/sudo
0       /run/console
0       /run/setrans
0       /run/unbound
0       /run/lvm
12K     /run/mdadm
4.0K    /run/tmpfiles.d
356K    /run/udev
0       /run/mount
1.2G    /run/log
0       /run/lock
4.0K    /run/initramfs
99M     /run/systemd
1.3G    /run

Code (Text):

du -h --max-depth=1 /run/log
1.2G    /run/log/journal
1.2G    /run/log

Code (Text):

df -hT
Filesystem     Type      Size  Used Avail Use% Mounted on
/dev/md1       ext4       69G   51G   15G  78% /
devtmpfs       devtmpfs   16G     0   16G   0% /dev
tmpfs          tmpfs      16G     0   16G   0% /dev/shm
tmpfs          tmpfs      16G  1.3G   15G   8% /run
tmpfs          tmpfs      16G     0   16G   0% /sys/fs/cgroup
tmpfs          tmpfs      16G     0   16G   0% /tmp
/dev/md2       ext4      151G  9.4G  134G   7% /home
tmpfs          tmpfs     3.2G     0  3.2G   0% /run/user/0
tmpfs          tmpfs     3.2G     0  3.2G   0% /run/user/992

After reboot

Code (Text):

du -h --max-depth=1 /run/systemd/
0       /run/systemd/shutdown
4.0K    /run/systemd/seats
0       /run/systemd/machines
4.0K    /run/systemd/users
4.0K    /run/systemd/sessions
0       /run/systemd/netif
0       /run/systemd/initctl
4.0K    /run/systemd/readahead
24K     /run/systemd/generator.late
20K     /run/systemd/generator
4.0K    /run/systemd/journal
0       /run/systemd/ask-password
0       /run/systemd/inaccessible
48K     /run/systemd/system
112K    /run/systemd/

buik · Dec 25, 2017

I don't like systemd at all.
Always been unstable with lots of (in)stability problems.

Arguments against systemd:

Breaking promises, immaturity, and (in)stability

Scope creep

Absurd bugs and responses

Conceptional problems

Scope creep leads to vulnerabilities

Poor design

Ignorance of fundamental operating system concepts

Seems hard for the EL7 team to get it stable.

Systemd is updated to a new version: rpms/systemd.git - git.centos.org
Even if it is against EL own philosophy to release bugfix and backport code instead of new software.
What is backporting and how does it affect Red Hat Enterprise Linux (RHEL)? - Red Hat Customer Portal

Moreover. Key members that developed systemd, are the software engineers working for the same Red Hat.

Andy · Dec 25, 2017

How do you reboot a dedicated server with reliablesite.net?
Couldn't find any info on their website so i sent them a ticket but hopefully anyone here using their service can have some input.

pamamolf · Dec 25, 2017

Just login to ssh and use the command: reboot

or do you mean something else?

eva2000 · Dec 25, 2017

yeah just reboot via ssh command line

as troublesome as systemd is, still prefer CentOS 7.x over 6.x

buik · Dec 25, 2017

Andy said: ↑

How do you reboot a dedicated server with reliablesite.net?
Couldn't find any info on their website so i sent them a ticket but hopefully anyone here using their service can have some input.
Click to expand...

Beside SSH, via the controle panel (every noteworthy provider has one),
IPMI or equivalent.

Log in / Register

Forums

Learn about Centmin Mod LEMP Stack today

systemd update

Matt Well-Known Member

pamamolf Well-Known Member

eva2000 Administrator Staff Member

buik “The best traveler is one without a camera.”

Andy Active Member

pamamolf Well-Known Member

eva2000 Administrator Staff Member

buik “The best traveler is one without a camera.”

.

Log in / Register

Useful Searches

Learn about Centmin Mod LEMP Stack today

systemd update

Matt Well-Known Member

pamamolf Well-Known Member

eva2000 Administrator Staff Member

buik “The best traveler is one without a camera.”

Andy Active Member

pamamolf Well-Known Member

eva2000 Administrator Staff Member

buik “The best traveler is one without a camera.”

.