Bug 51506 - RFE: robustify sshd restart with sshd -t
Summary: RFE: robustify sshd restart with sshd -t
Keywords:
Status: CLOSED DEFERRED
Alias: None
Product: Red Hat Linux
Classification: Retired
Component: openssh
Version: 7.3
Hardware: i386
OS: Linux
medium
medium
Target Milestone: ---
Assignee: Nalin Dahyabhai
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2001-08-11 11:07 UTC by Pekka Savola
Modified: 2008-05-01 15:38 UTC (History)
0 users

Fixed In Version:
Doc Type: Enhancement
Doc Text:
Clone Of:
Environment:
Last Closed: 2001-08-11 11:07:18 UTC
Embargoed:


Attachments (Terms of Use)

Description Pekka Savola 2001-08-11 11:07:13 UTC
There have been some problems with 'sshd restart' etc. especially if config file 
is wrong (e.g. outdated sshd_config won't start with new sshd).  There are a few bugzilla 
reports on this.

If this something direly in need of getting more robust, you could consider applying:

20010806
 - OpenBSD CVS Sync 
   - stevesk.org 2001/07/26 17:18:22
     [sshd.8 sshd.c]
     add -t option to test configuration file and keys; pekkas
     ok markus@

Running 'sshd -t' before 'restart' and 'condrestart' in init.d/sshd script, and requiring it will pass 
before doing restart might reduce the cases where your connection is killed if new sshd is not restarted
due to some error.

(it's pretty late for this, and I'm not sure if this is viewed as something very important, but here it is anyway...)

Comment 1 Nalin Dahyabhai 2001-09-06 12:50:35 UTC
Doing so would only postpone the startup failure to next-boot, and also
potentially break from the restart-on-upgrade behavior most packages exchibit. 
As I have no strong feeling either way about this, I'm inclined to mark this bug
"deferred" and just go with what the upstream init script will do once the patch
works its way into a release.

If you believe that this can't wait, please reopen this bug.


Note You need to log in before you can comment on or make changes to this bug.