[solved] ssh went down after recent binary update

wolfdogg · 2013-07-20 00:25:04

well i just did a binary update and ran into some problems apparently not knowing what i was doing, see last couple posts here https://bbs.archlinux.org/viewtopic.php … 3#p1300263. So now i am trying to reinstall some of my apps, but when i go to ssh into it i get connection activily refused. so i go to the console, and run

# systemctl restart sshd

no output message. then i ran systemctl stop sshd, start sshd, etc... no output messages. then i get

# ps aux | grep sshd
root 202 0.0 0.0 39736 2684 ? Ss 09:20 0:00 /usr/bin/sshd -D

then i run nmap on port 22 localhost and get port 22/tcp state closed, host is up.

Note, pasting my conf file will be impossible without ssh. catch 22. I didnt make any edits to that ssh conf file in more than 6 months, all has been fine until the binary update. Should i be looking at that as the cause or is there something deeper going on here?

Edit;

 # journalctl -u sshd
Starting OpenSSH Daemon
Started OpenSSH Daemon
Server listening on :: port 22.
Server listening on 0.0.0.0 port 22

doesnt look very secure but apart from that looks fine to me atleast.

Last edited by wolfdogg (2013-09-30 21:17:31)

wolfdogg · 2013-07-20 00:42:35

It looks like i was having a problem with network issues, which is very wierd, because i have netcfg setup to automatically to load two networks, (1 gigabit direct connect, and one wan/lan to a router) and havent touched it since well before binary updates. Once i brought all connections down and upped only the pertinent one, im good. I even rebooted the system before that which should have solved it if it was a temporary issue. So something must have changed in the settings.

I have been using systemd for months now, but found recently that i never uninstalled initscripts or installed sysvcompat , so I took care of that right after i updated the binaries a couple days ago just before i noticed this problem . What could this have done to kill SSH or the network? ping still works in and out. Note, i wasnt depending on anything in the rc.conf file as far as network, as far as i know i moved all that stuff off months ago. i can get a paste here what was remaining before i unistalled initscripts.

  1 #
  2 # /etc/rc.conf - Main Configuration for Arch Linux
  3 #
  4
 24 DAEMON_LOCALE="no"
 25 HARDWARECLOCK="localtime"
 26 #TIMEZONE="America/Los_Angeles"
 27 KEYMAP="us"
 28 CONSOLEFONT=
 29 CONSOLEMAP=
 30 USECOLOR="yes"
 31
 42 MODULES=(fuse)
 43
 44 # Udev settle timeout (default to 30)
 45 UDEV_TIMEOUT=30
 46
 47 # Scan for FakeRAID (dmraid) Volumes at startup
 48 USEDMRAID="no"
 49
 50 # Scan for BTRFS volumes at startup
 51 USEBTRFS="no"
 52
 53 # Scan for LVM volume groups at startup, required if you use LVM
 54 #USELVM="no"
 55 USELVM="yes"
 56
 57 # -----------------------------------------------------------------------
 58 # NETWORKING
 63 HOSTNAME="falcon"
 80
 81 # -----------------------------------------------------------------------
 82 # DAEMONS
 83 # -----------------------------------------------------------------------
 93 DAEMONS=(hwclock syslog-ng dbus !network zfs-fuse netfs xinetd crond samba httpd sshd @mysqld @postfix)

So this means each reboot i will have to muck with netcfg until i get this figured out.

Last edited by wolfdogg (2013-07-20 00:50:11)

jasonwryan · 2013-07-20 01:15:49

How did you restart sshd without ssh access?

wolfdogg · 2013-07-20 03:59:32

from the console (the arch box itself), i have two computers right next to each other in this case. whatever it was, it must be a fluke, cause i just restarted again after bringing down and back up the network manually, and the network is now connecting itself without intervention again as it should be through systemd netcfg. Wierd since i was able to ping that box, and from that box i was able to ping the machine that i was trying to ssh into it with. All signs pointed to it being able to work, but connection was actively being refused.

Ill post back if it gives me a hassle but are you seeing anything on my old rc.conf that might have been somehow being initiated before i uninstalled initscripts, even though systemd was being used? If its not that, then it must be my network equipment, i.e. my consumer level router and switch equipment. .

Last edited by wolfdogg (2013-07-20 04:06:40)

wolfdogg · 2013-07-21 23:15:21

yeah netcfg is regularly giving problems loading one of two connections

[root@falcon Backup Set 2013-05-06 134944]# journalctl | grep netcfg
Jul 20 11:26:44 falcon netcfg[8102]: :: wolfnet down rm: cannot remove ‘/run/network/profiles/wolfnet’: No such file or directory
Jul 20 11:26:44 falcon netcfg[8102]: [done]
Jul 20 11:26:44 falcon netcfg-daemon[8103]: :: wolfnet down [done]
Jul 20 11:26:44 falcon netcfg-daemon[8103]: :: hispeednet down [done]
Jul 20 14:05:11 falcon netcfg-daemon[255]: :: wolfnet up :: hispeednet up [done]
Jul 20 14:05:11 falcon netcfg-daemon[255]: RTNETLINK answers: File exists
Jul 20 14:05:11 falcon netcfg-daemon[255]: Could not configure interface
Jul 20 14:05:11 falcon netcfg-daemon[255]: [fail]
Jul 20 14:05:11 falcon netcfg[254]: :: wolfnet up [done]
Jul 20 14:25:28 falcon netcfg[872]: :: wolfnet down [done]
Jul 20 14:25:28 falcon netcfg-daemon[873]: :: hispeednet down [done]
Jul 20 14:27:38 falcon netcfg[240]: :: wolfnet up [done]
Jul 20 14:27:38 falcon netcfg-daemon[241]: :: hispeednet up [done]
Jul 20 14:47:26 falcon netcfg[458]: :: wolfnet down [done]
Jul 20 14:47:26 falcon netcfg-daemon[459]: :: wolfnet down [done]
Jul 20 14:47:26 falcon netcfg-daemon[459]: :: hispeednet down [done]
Jul 20 15:03:39 falcon netcfg[252]: :: wolfnet up RTNETLINK answers: File exists
Jul 20 15:03:39 falcon netcfg[252]: Could not configure interface
Jul 20 15:03:39 falcon netcfg[252]: [fail]
Jul 20 15:03:39 falcon netcfg-daemon[253]: :: wolfnet up :: hispeednet up [done]
Jul 20 15:03:39 falcon netcfg-daemon[253]: [done]
Jul 20 15:03:39 falcon systemd[1]: Unit netcfg@wolfnet.service entered failed state.
Jul 21 03:04:10 falcon netcfg-daemon[1819]: :: wolfnet down [done]
Jul 21 03:04:10 falcon netcfg-daemon[1819]: :: hispeednet down [done]
Jul 21 03:34:08 falcon netcfg-daemon[376]: :: hispeednet up :: wolfnet up [done]
Jul 21 03:34:08 falcon netcfg-daemon[376]: [done]
Jul 21 03:34:08 falcon netcfg[374]: :: wolfnet up RTNETLINK answers: File exists
Jul 21 03:34:08 falcon netcfg[374]: Could not configure interface
Jul 21 03:34:08 falcon netcfg[374]: [fail]
Jul 21 03:34:08 falcon systemd[1]: Unit netcfg@wolfnet.service entered failed state.
Jul 21 03:51:27 falcon netcfg-daemon[673]: :: wolfnet down [done]
Jul 21 03:51:29 falcon netcfg-daemon[673]: :: hispeednet down [done]
Jul 21 03:53:01 falcon netcfg[367]: mkdir: cannot create directory ‘/run/network/suspend’: File exists
Jul 21 03:53:01 falcon netcfg[367]: :: wolfnet up [done]
Jul 21 03:53:01 falcon netcfg-daemon[369]: :: hispeednet up :: wolfnet up [done]
Jul 21 03:53:01 falcon netcfg-daemon[369]: RTNETLINK answers: File exists
Jul 21 03:53:01 falcon netcfg-daemon[369]: Could not configure interface
Jul 21 03:53:01 falcon netcfg-daemon[369]: [fail]
Jul 21 04:50:01 falcon netcfg[734]: :: wolfnet down [done]
Jul 21 04:50:01 falcon netcfg-daemon[735]: :: hispeednet down [done]
Jul 21 04:53:05 falcon netcfg-daemon[370]: :: wolfnet up :: hispeednet up [done]
Jul 21 04:53:05 falcon netcfg-daemon[370]: [done]
Jul 21 04:53:05 falcon netcfg[368]: :: wolfnet up RTNETLINK answers: File exists
Jul 21 04:53:05 falcon netcfg[368]: Could not configure interface
Jul 21 04:53:05 falcon netcfg[368]: [fail]
Jul 21 04:53:05 falcon systemd[1]: Unit netcfg@wolfnet.service entered failed state.

i probably have something misconfigured, network has always been an uphill battle.

Last edited by wolfdogg (2013-07-21 23:16:37)

wolfdogg · 2013-09-30 21:17:19

this problem may have been due to no freespace on the drive, see this post https://bbs.archlinux.org/viewtopic.php?id=169985 for reference. it was affecting network, samba, and mysql. problem was solved by using gparted to pull some freespace from /home and put it in root, and i upgraded to netctl.

Arch Linux

#1 2013-07-20 00:25:04

[solved] ssh went down after recent binary update

#2 2013-07-20 00:42:35

Re: [solved] ssh went down after recent binary update

#3 2013-07-20 01:15:49

Re: [solved] ssh went down after recent binary update

#4 2013-07-20 03:59:32

Re: [solved] ssh went down after recent binary update

#5 2013-07-21 23:15:21

Re: [solved] ssh went down after recent binary update

#6 2013-09-30 21:17:19

Re: [solved] ssh went down after recent binary update

Board footer