This is a mitigation for a buffer overflow in Solaris' PAM username
handling (CVE-2020-14871), and is only enabled for Sun-derived PAM
implementations. This is not a problem in sshd itself, it only
prevents sshd from being used as a vector to attack Solaris' PAM.
It does not prevent the bug in PAM from being exploited via some other
PAM application.
Based on github PR#212 from Mike Scott but implemented slightly
differently. ok tim@ djm@
sigaction(2). This wrapper blocks all other signals during the handler
preventing races between handlers, and sets SA_RESTART which should reduce
the potential for short read/write operations.
OpenBSD-Commit-ID: 5e047663fd77a40d7b07bdabe68529df51fd2519
According to POSIX, pthread_create returns the failure reason in
the non-zero function return code so make the fork wrapper do that.
Matches previous change.
Some platforms such as HP-UX do not have pam_putenv. Currently the
calls are ifdef'ed out, but a new one was recently added. Remove the
ifdefs and add a no-op implementation. bz#3008, ok djm.
remove the legacy one.
Includes a fairly big refactor of auth2-pubkey.c to retain less state
between key file lines.
feedback and ok markus@
OpenBSD-Commit-ID: dece6cae0f47751b9892080eb13d6625599573df
bz#2249: handle the case where PAM returns PAM_MAXTRIES by ceasing to offer
password and keyboard-interative authentication methods. Should prevent
"sshd ignoring max retries" warnings in the log. ok djm@
It probably won't trigger with keyboard-interactive in the default
configuration because the retry counter is stored in module-private
storage which goes away with the sshd PAM process (see bz#688). On the
other hand, those cases probably won't log a warning either.
When sshd decides to not allow a login (eg PermitRootLogin=no) and
it's using PAM, it sends a fake password to PAM so that the timing for
the failure is not noticeably different whether or not the password
is correct. This behaviour can be detected by sending a very long
password string which is slower to hash than the fake password.
Mitigate by constructing an invalid password that is the same length
as the one from the client and thus takes the same time to hash.
Diff from djm@
[canohost.c channels.c cipher-acss.c defines.h dns.c gss-genr.c]
[gss-serv-krb5.c gss-serv.c log.h loginrec.c logintest.c readconf.c]
[servconf.c ssh-keygen.c ssh-keyscan.c ssh-keysign.c ssh-rand-helper.c]
[ssh.c sshconnect.c sshd.c openbsd-compat/bindresvport.c]
[openbsd-compat/bsd-arc4random.c openbsd-compat/bsd-misc.c]
[openbsd-compat/getrrsetbyname.c openbsd-compat/glob.c]
[openbsd-compat/mktemp.c openbsd-compat/port-linux.c]
[openbsd-compat/port-tun.c openbsd-compat/readpassphrase.c]
[openbsd-compat/setproctitle.c openbsd-compat/xmmap.c]
make the portable tree compile again - sprinkle unistd.h and string.h
back in. Don't redefine __unused, as it turned out to be used in
headers on Linux, and replace its use in auth-pam.c with ARGSUSED
session.c ssh-rand-helper.c sshd.c openbsd-compat/bsd-cygwin_util.c
openbsd-compat/setproctitle.c] Convert malloc(foo*bar) -> calloc(foo,bar)
in Portable-only code; since calloc zeros, remove now-redundant memsets.
Also add a couple of sanity checks. With & ok djm@
[buffer.c channels.c deattack.c misc.c scp.c session.c sftp-client.c]
[sftp-server.c ssh-agent.c ssh-rsa.c xmalloc.c xmalloc.h auth-pam.c]
[uidswap.c]
change OpenSSH's xrealloc() function from being xrealloc(p, new_size)
to xrealloc(p, new_nmemb, new_itemsize).
realloc is particularly prone to integer overflows because it is
almost always allocating "n * size" bytes, so this is a far safer
API; ok deraadt@
