[SOLVED] Can't ssh after upgrade

[mpadilla2]

I'm working on a friend's installation. I upgraded the alarm installation. Made sure that systemd-networkd was enabled and that I had a eth0.network file in /etc/systemd/network with the proper information. Everything went smoothly. I can ping the ip and get a reply. However, I am no longer able to ssh back into it. Here is what I get:

$this->bbcode_second_pass_code('', 'ssh -vvvvv -p 25300 shatewie@192.168.1.253
OpenSSH_6.0p1 Debian-4+deb7u2, OpenSSL 1.0.1j 15 Oct 2014
debug1: Reading configuration data /etc/ssh/ssh_config
debug1: /etc/ssh/ssh_config line 19: Applying options for *
debug2: ssh_connect: needpriv 0
debug1: Connecting to 192.168.1.253 [192.168.1.253] port 25300.
debug1: connect to address 192.168.1.253 port 25300: Connection refused
ssh: connect to host 192.168.1.253 port 25300: Connection refused
')

I've tried clearing known_hosts on the client. I have tried attaching the drive on another plug e02, chrooting in and enabling sshd.socket and running ssh-keygen but nothing has helped. I ran systemctl list-unit-files while chrooted and sshd.socket is enabled. I did change the ListenStream setting in the unit file to 25300. I'm putting together a serial cable to try that next. Before I try that does anyone have any other suggestions? I'm sure there is a way to get this back just not sure what to try next.

Edit: the router has 25300 port forwarded properly

[pepedog]

Which armv5 is this, and what kernel version?
Don't assume same IP address allocated, btw

[mpadilla2]

I set up a static ip so it is the same ip. As I said I am able to ping it and it replies so that part is working. Here is my working static ip setup for systemd-networkd:

$this->bbcode_second_pass_code('', '[Match]
Name=eth0

[Network]
Address=192.168.1.253/24
Gateway=192.168.1.2
DNS=8.8.8.8
')

Any suggestions on the nonworking ssh?

[WarheadsSE]

Did you actually change the port to 25300?

[mpadilla2]

Update:
today I chrooted in again while the drive was connected to another working alarm on pogo e02. I enabled telnet.socket. Reinstalled the drive to the other pogo and started up. Attemped to telnet in and get connection refused. So it seems something is preventing telnet and ssh from starting up. Researching further....but any suggestions at this point would be greatly appreciated.

[mpadilla2]

sshd_config is configured to listen on port 25300. It was configured this way before the upgrade and worked fine. Only after the upgrade it is not working. But telnet is also not working so something is preventing both from starting up it seems. Any ideas?

[mpadilla2]

Figured it out

I chrooted in and ran journalctl.

The log showed it was expecting to mount the second hard drive. When that hard drive was not found it would fail and shutdown all services. I commented out the second drive in /etc/fstab and started it up. I was then able to ssh in successfully.

My question is why such a catastrophic failure just because the second drive is missing? It is only storage not a rootfs.

[Socaltom]

Yes,
That's a problem I've run into before.
Tom

[pepedog]

Perhaps option nofail in fstab might help?

[mpadilla2]

Thanks pepedog! Added it.