Bug #5314: /etc/sshd script fails to restart SSH properly - pfSense - pfSense bugtracker

Actions

Copy link

Bug #5314

closed

/etc/sshd script fails to restart SSH properly

Added by Kill Bill over 9 years ago. Updated about 9 years ago.

Status:

Closed

Priority:

Normal

Assignee:

Category:

Target version:

Start date:

10/17/2015

Due date:

% Done:

Estimated time:

Plus Target Version:

Release Notes:

Affected Version:

All

Affected Architecture:

All

Description

I pretty much only noticed this due to sshdcond package which tries to use the script on package resync. This more often than not results in something along these lines:

Oct 17 10:20:33    sshd[7250]: Received signal 15; terminating.
Oct 17 10:20:33    sshd[45658]: fatal: Cannot bind any address.
Oct 17 10:20:33    sshd[45658]: fatal: Cannot bind any address.
Oct 17 10:20:33    sshd[45658]: error: Bind to port 22 on 0.0.0.0 failed: Address already in use.
Oct 17 10:20:33    sshd[45658]: error: Bind to port 22 on 0.0.0.0 failed: Address already in use.
Oct 17 10:20:33    sshd[45658]: error: Bind to port 22 on :: failed: Address already in use.
Oct 17 10:20:33    sshd[45658]: error: Bind to port 22 on :: failed: Address already in use.

Actions

Copy link

Updated by Kill Bill over 9 years ago

Here's a second part of that log, showing that the bug ultimately results in SSH being killed instead of restarted. :-(

Oct 17 10:21:01    sshd[61892]: Server listening on 0.0.0.0 port 22.
Oct 17 10:21:01    sshd[61892]: Server listening on :: port 22.
Oct 17 10:21:00    check_reload_status: starting sshd
Oct 17 10:21:00    php: servicewatchdog_cron.php: Service Watchdog detected service sshd stopped. Restarting sshd (Secure Shell Daemon)

So, if it's not added to the Service Watchdog package, you end up cut off SSH access.

Actions

Copy link

Updated by Chris Buechler over 9 years ago

Status changed from New to Feedback
Affected Version changed from 2.2.x to All

there something atypical that triggers that? It appears the sigterm doesn't complete before starting it again, but not able to replicate that on anything, full or nano, after numerous attempts. Slowest thing I have handy at the moment is an APU.

Actions

Copy link