similar to: Okay...

I'm having trouble getting any sort of work-around for 3.10 on Solaris 8 with Sun's tcsh. I've tried using "hup" to correct it but to no avail. This problem wasn't present with ssh version 1 - it just seem to work. Now we get all kinds of abandoned ssh processes lying around that have to be manually killed. Does anyone know if there is going to be a fix for this problem

I ran into a weird problem yesterday where Putty can't connect to my OpenSSH/Sol box (and still can't ever since). On the Putty side I get the error message: "internal fault: chaos in SSH 2 transport layer" I've attached the "sshd -d -d -d" syslog output. Any ideas what's going wrong here? From a layman's point of view, it would appear that Putty and

It appears that "fixpaths" has "/usr/bin/perl" hard-coded in. This causes make to fail immediately after running configure. Ed Ed Phillips <ed at udel.edu> University of Delaware (302) 831-6082 Systems Programmer III, Network and Systems Services finger -l ed at polycut.nss.udel.edu for PGP public key

Hi, Sorry no core dumps, the sshd programme is perfectly happy just fails to consider changing the password. Cheers, Martyn -----Original Message----- From: Ed Phillips [mailto:ed at UDel.Edu] Sent: 06 November 2001 18:38 To: Roberts,M,Martyn,IVLH4 C Cc: openssh-unix-dev Subject: Re: Solaris 7 changing password via PAM On Tue, 6 Nov 2001 martyn.a.roberts at bt.com wrote: > Date: Tue, 6

I'm guess I wasn't following the whole cookies discussion completely (putting cookies in /tmp to avoid putting them on NFS, etc.), but I noticed today that with 2.9.9p2, if I use "ssh -X" to start a shell on the server, in that shell XAUTHORITY is set to /tmp/ssh-XXXXXXXX/cookies and there are cookies placed there there. These are the "fake" cookies for the

I don't know if this is still a problem in the latest snapshot, but with 2.9.9p2, if you do a "./configure ... --with-tcp-wrappers", there's no way to specify a location for tcpd.h and libwrap.a. This is troublesome on Solaris where you might install stuff like that in /opt/lib or /usr/local/lib or something that is not searched by default. Ed Ed Phillips <ed at

If I change openbsd-compt/readpassphrase.c and type "make" from the top-level, nothing happens. I have to remove openbsd-compat/libopenbsd-compat.a to get "make" to do its thing. Ed Phillips <ed at udel.edu> University of Delaware (302) 831-6082 Systems Programmer III, Network and Systems Services finger -l ed at polycut.nss.udel.edu for PGP public key

Can someone tell be briefly how to get OpenSSH3.1p1 configured and compiled to use /dev/random? Can OpenSSH use /dev/random directly now? Thanks, Ed Ed Phillips <ed at udel.edu> University of Delaware (302) 831-6082 Systems Programmer III, Network and Systems Services finger -l ed at polycut.nss.udel.edu for PGP public key

On Mon, 8 Oct 2001, Markus Friedl wrote: > Date: Mon, 8 Oct 2001 23:12:41 +0200 > From: Markus Friedl <markus at openbsd.org> > To: Ed Phillips <ed at UDel.Edu> > Subject: Re: ssh hangs instead of exiting > > On Mon, Oct 08, 2001 at 05:07:12PM -0400, Ed Phillips wrote: > > On Mon, 8 Oct 2001, Markus Friedl wrote: > > > > > Date: Mon, 8 Oct 2001

In do_pam_cleanup_proc(), there are 3 calls to PAM: 1) pam_close_session() - do lastlog stuff 2) pam_setcred(PAM_DELETE_CRED) - delete credentials 3) pam_end() - close PAM It appears that pam_setcred() always fails with the error PAM_PERM_DENIED. This is due to a check done pam_unix.so to not allow a caller with euid 0 to even try to delete their SECURE_RPC credentials. When sshd calls

The reason I ask about the patches is because I think the problem you're seeing might actually be a bug in pam_unix.so.1 - it's something to try at least. We don't use password aging and we don't use the "passwd" command to change passwords, so we haven't run into this at our site even though we probably don't have pam_unix.so patched up. Also, the passwd

I've been doing so more tests with 2.9.9p2 on Sol8. Here are my finding so far: When a user needs to change his password and trys to run a command in non-interactive mode, it just succeeds without even trying to prompt the user for a new password. Damien submitted a fix - it works for me (is it going into CVS?). When a user needs to change his password and trys to login in interactive

(I vaguely remember talk about PAM session stuff recently... please excuse me if this is the same problem.) I compiled v2.9.9p2 on Solaris 8 with the following configuration and the Sun Workshop v5 compiler: OpenSSH has been configured with the following options: User binaries: /opt/openssh-2.9.9p2/bin System binaries: /opt/openssh-2.9.9p2/sbin

If I set it to "no", should I still be able to login with a typed-in password? I get messages such as these: --- Jul 2 12:23:39 remedy.udel.edu sshd[6811]: [ID 800047 local4.debug] debug1: userauth-request for user ed service ssh-connection method password Jul 2 12:23:39 remedy.udel.edu sshd[6811]: [ID 800047 local4.debug] debug1: attempt 1 failures 1 Jul 2 12:23:39 remedy.udel.edu

Is anyone working on a patch for this feature? I don't remember seeing anyone say whether they were actually going to try to implement it in the current OpenSSH code... Thanks, Ed Ed Phillips <ed at udel.edu> University of Delaware (302) 831-6082 Systems Programmer III, Network and Systems Services finger -l ed at polycut.nss.udel.edu for PGP public key

I tried replacing readpassphrase() for v2.9.9p2 on Sol8 with a different version that just calls getpassphrase(). It appears to solve the echo problem when the user tries to login in interactive mode and needs to change their password. Can anyone else try this with v2.9.9p2 on Solaris? Be sure to add: #define HAVE_GETPASSPHRASE ... to config.h when compiling (since it's not a configurable

When I run a command like on v2.9.9p2 on Solaris 8: client> ssh server ... and immediately press CTRL-D, I get a shell prompt from the remote host, but I can't type into it and have to kill the ssh process on the client. Ed Ed Phillips <ed at udel.edu> University of Delaware (302) 831-6082 Systems Programmer III, Network and Systems Services finger -l ed at polycut.nss.udel.edu

As it turns out, you don't need to configure OpenSSL 0.9.6c in any special way to have support for /dev/urandom - it automatically tries to use it if it exists on a Unix system. So, I compiled the current snapshot using "--with-rand-helper=no" and everything appears to build and work fine on Solaris 8. Ed Ed Phillips <ed at udel.edu> University of Delaware (302) 831-6082

cc -g -I. -I. -I/opt/openssl-0.9.6c/include -I. -I/usr/local/include -DETCDIR=\"/opt/openssh-3.0.2p1-x509/etc\" -D_PATH_SSH_PROGRAM=\"/opt/openssh-3.0.2p1-x509/bin/ssh\" -D_PATH_SSH_ASKPASS_DEFAULT=\"/opt/openssh-3.0.2p1-x509/libexec/ssh-askpass\" -D_PATH_SFTP_SERVER=\"/opt/openssh-3.0.2p1-x509/libexec/sftp-server\" -D_PATH_SSH_PIDDIR=\"/var/run\"

Okay, I'm confused again. They way you guys are talking about the conversation routine, it would seem that you think it is a way to fetch something from the user - like a new password. Is this possible? Does calling pam_chauthtok() cause the underlying pam_sm_chauthtok() eventually print something on stdout and read a new password from stdin (the socket to the client) using the conversation