similar to: Fwd: Re: Bus Error with OpenSSH 3.7.1p2 on Solaris 8, SPARC 64-bit

I had a problem much like the one that Matthias Koeppe experienced. I would only get the BUS error when running in a YASSP-modified system, though. The binary worked fine on a non-YASSP host. I applied his change to use a long for the mask, and it works now. Cheers, -Thomas __________________________________ Do you Yahoo!? The New Yahoo! Shopping - with improved product search

The following patch appears to fix the BUS error received on Solaris 8. This problem manifests as an immediate disconnect with no apparent cause immediately after authentication with the host. --- Darren Tucker <dtucker at zip.com.au> wrote: > Date: Tue, 30 Sep 2003 09:35:26 +1000 > From: Darren Tucker <dtucker at zip.com.au> > Subject: Re: Bus Error with OpenSSH 3.7.1p2 on

I built OpenSSH as a 64-bit binary on Solaris 9, using gcc 3.3.2, OpenSSL 0.9.7c and zlib 1.2.1. sshd starts up normally, and will begin the login session, however, some time after it reads /etc/default/login, it faults and kills the connection. There are no error messages logged to syslog (with LogLevel set to DEBUG) or with the -ddd option. Here's a truss of the failure: 8132:

openssh 3.7p1 sshd on Solaris 8 / sparcv9: sshd runs fine, and starts to allow the login. However, when reading from /etc/default/login, I get a bus error. I am able to get sshd to work by commenting out these lines in session.c: 1015,1018c1015 < # ifdef HAVE_ETC_DEFAULT_LOGIN < read_etc_default_login(&env, &envsize, pw->pw_uid); < path =

Hi, I know that with readlines you can start executing a code upon user input. What I would like to do is to have the ability to abort a code based on a new user input. E.g. when the user hits enter the code stops running. That is, I would periodically check for a user input and stop the code when the correct input was entered. Below is an example of a function which would run an endless loop

Can i somehow append objects to an .Rdata file? I didn't see an option for it in the save() method. dump() won't work since i have s4 objects in there.

I compiled openssh 3.7.1p1 on Solaris 9 with the Forte compiler in 64-bit mode. After authentication, a forked child of sshd dies with a Bus Error in `read_etc_default_login' (session.c). The reason is the use of `sscanf' with control string "%5lo" on a `mode_t' value. On Solaris in 64-bit mode, `mode_t' is an `unsigned int' (32 bits), whereas `long' is 64

I redid my previous OPIE patch for the current ssh tree. It seems to work fine here, and I'ld love to see it merged before the 3.0 release. Wichert. diff -x CVS -wNur ../cvs/other/openssh_cvs/Makefile.in openssh_cvs/Makefile.in --- ../cvs/other/openssh_cvs/Makefile.in Mon Oct 22 02:53:59 2001 +++ openssh_cvs/Makefile.in Sun Nov 4 01:18:19 2001 @@ -50,7 +50,7 @@ SSHOBJS= ssh.o

This is from the current CVS: % make ... In file included from config.h:629, from /home/mo/project/openssh_cvs/includes.h:20, from /home/mo/project/openssh_cvs/scp.c:77: /home/mo/project/openssh_cvs/defines.h:208: warning: redefinition of `clock_t' /usr/include/time.h:60: warning: `clock_t' previously declared here gcc -o scp scp.o -L.

Is anyone maintaining the OSF_SIA support in openssh? This seems to be an obvious bug triggered if you try to connect as a non-existant user. >From auth1.c line 459 #elif defined(HAVE_OSF_SIA) (sia_validate_user(NULL, saved_argc, saved_argv, get_canonical_hostname(), pw->pw_name, NULL, 0, NULL, "") == SIASUCCESS)) { #else /*

Hi, I have a problem with the extern declarations of optarg, optind, etc. We're currently moving getopt from being a statically linked function to a dynamically linked function as part of the Cygwin DLL. On Windows, this requires to generate special symbols (__imp__optarg, etc.), which is done by marking the exported variables in the corresponding header. Instead of extern char *optarg;

Hi All. While wandering in auth-pam.c I noticed that there's a few Portable-specific escapees from the xmalloc(foo * bar) cleanup. There's also a "probably can't happen" integer overflow in ssh-rand-helper.c with the memset: num_cmds = 64; - entcmd = xmalloc(num_cmds * sizeof(entropy_cmd_t)); + entcmd = xcalloc(num_cmds, sizeof(entropy_cmd_t));

Dear All, I can install OpenSSH 3.7.1p2 with OpenSSL 0.9.7c on HP-UX 11.0 without problem. However, I find that all valid accounts are treated as "locked" (even specify "AllowUsers USER1" in sshd_config) and can't make SSH connection to the server. Here is the message shown in syslog.log. Oct 3 16:49:17 SERVER_NAME sshd[12994]: User USER1 not allowed because account is

Sorry if this is stuff that's been talked about before. If it is, just ignore me. I'm curious to know why Portable OpenSSH doesn't include @bindir@ in the _PATH_STDPATH. This would save most installers of portable OpenSSH from having to --with-default-path=$PREFIX/bin in order to ensure that scp will work properly. This would also, I imagine, save quite a lot of hassle for

sorry, Darren. Long over due comments. [..] >+/* Record a failed login attempt. */ >+void >+record_failed_login(const char *user, const char *host, const char *ttyname) >+{ >+#ifdef WITH_AIXAUTHENTICATE >+ loginfailed(user, host, ttyname); >+#endif >+#ifdef _UNICOS >+ cray_login_failure((char *)user, IA_UDBERR); >+#endif /* _UNICOS */ >+} I like the

what do you think about this patch to add the path to scp to _PATH_STDPATH? is there a better or cleaner way to do this? i'm hoping to ward off 'scp doesn't work' questions for the next release. i did *not* add this to a --with-default-path path, because if a user specifies that, they should control its value completely. Index: Makefile.in

Hi, the SETGROUPS_NOOP define is used only for Cygwin. But it's also combined with the HAVE_SETGROUPS test in bsd-misc.c. So, since setgroups() exists and is functional in Cygwin, this is actually a noop in the more general sense. The below patch drops that entirely. Corinna Index: acconfig.h =================================================================== RCS file:

Hi, attached is a patch which solves the following problem: Sometimes scp could hang in Cygwin when used as remote end using the -t option. This is due to a binmode/textmode problem which could be raised by the login shell which is used by the user and it's setting of textmode on stdin. The patch solves that problem by explicitly setting binmode on stdin. Besides solving the hanging

I experienced a problem compiling openssh_cvs with IRIX 6.5.11f. % cc -version MIPSpro Compilers: Version 7.3.1.2m % cc -n32 -mips3 -O2 -OPT:Olimit=0 ..... -c auth.c cc-1515 cc: ERROR File = auth.c, Line = 100 A value of type "int" cannot be assigned to an entity of type "char *". while ((cp = getusershell()) != NULL) ^ 1 error detected in the

Hi, I've sent that patch once already but it seems more or less forgotten in the tumultuous days of the latest vulnerability. It adds a new define NO_IPPORT_RESERVED_CONCEPT which can be defined on platforms not supporting the concept of "privileged" ports only accessible by privileged users but which allow everyone to use these ports. This patch removes some Cygwin dependencies