search for: ssh_prng_command

Howdy, I recently had a problem with one of our servers (crashed due to power failure :-). While this shouldn't have been a problem for most of the workstations and servers on the network I noticed that I wasn't able to use ssh anymore. Ssh would simply hang during the connection. rsh and telnet however were able to connect without problem so there was no problem with the destination or

...nd it works > just fine. I found some time debugging the problem with using the built-in entropy generator in openssh-2.1.0. In the current condition, it should crash on every reasonable (?) system. The problem is in entropy.c in a debug call with a string argument (%s) of 0, if the command in ssh_prng_commands has no arg. Interesting enough, there is a "#if 1" in front of the offending statement and there is a note "FIXME: turn this off later", too. Why turn it off, it should work now and is a useful information in debug mode. Patch is following here - it works for me and my ssh_prn...

It would be very useful if openssh could have the same sort of hostinstall target that traditional ssh does. The ssh programs only have to be installed once, for central distribution via nfs or rdist or whatever, but every host needs its key ... Thanks for a great program. karl at gnu.org

Hi, I'm running openssh 2.1.1p4 on Solaris 7 (sparc). Occationally, when I boot up the server, the startup script I wrote to start sshd fails to start sshd with the following error: fatal: Not enough entropy in RNG What am I doing wrong?? Is there anything I can do to prevent this from happening? Is just restarting sshd a valid thing to do?? Thanks for any thoughts, David