search for: main_sigchld_handl

Hi What if you have sysv signals (i.e. signal is restored when handler is called) and child process exits here? Zombie will be left, because SIGCHLD is ignored at that point. Shouldn't signal be before waitpid? Mikulas static void main_sigchld_handler(int sig) { int save_errno = errno; pid_t pid; int status; while ((pid = waitpid(-1, &status, WNOHANG)) > 0 || (pid < 0 && errno == EINTR)) ; <----- HERE signal(SIGCHLD, main_sigchld_handler); err...

...error(errno)); return 0.0; + } } RAND_add(&status, sizeof(&status), 0.0); diff -r -c openssh-3.1p1/sshd.c openssh-3.1p1-peter/sshd.c *** openssh-3.1p1/sshd.c Tue Mar 5 02:31:30 2002 --- openssh-3.1p1-peter/sshd.c Mon Mar 18 22:03:37 2002 *************** *** 264,273 **** main_sigchld_handler(int sig) { int save_errno = errno; ! int status; ! while (waitpid(-1, &status, WNOHANG) > 0) ! ; signal(SIGCHLD, main_sigchld_handler); errno = save_errno; --- 264,274 ---- main_sigchld_handler(int sig) { int save_errno = errno; ! int status, code; ! ! wh...

On Fri, 7 Feb 2020 at 15:09, Hisashi T Fujinaka <htodd at twofifty.com> wrote: > > On Fri, 7 Feb 2020, Damien Miller wrote: > > > On Thu, 6 Feb 2020, Hisashi T Fujinaka wrote: > > > >> Built and passed tests on NetBSD-9_RC2 on amd64 and on NetBSD-current on > >> amd64. Issues with MacOS Catalina: > >> configure: error: *** working libcrypto not

...passed on to the openbsd people. There is a related problem I noticed when doing this. If I open a bunch of connections to ssh (say in a perl script) and then close them all at once only some of the children get reaped on the server. The reason for this seems to be that in the SIGCHLD handler, main_sigchld_handler, uses a wait call to reap one child. If you have multiple children dying at roughly the same time some do not get caught by the handler and thus not reaped. They remain as zombies until sshd is restarted (at which time the get reaped). To circumvent this I have replaced the wait call with a l...

...error(errno)); return 0.0; + } } RAND_add(&status, sizeof(&status), 0.0); diff -r -c openssh-3.1p1/sshd.c openssh-3.1p1-peter/sshd.c *** openssh-3.1p1/sshd.c Tue Mar 5 02:31:30 2002 --- openssh-3.1p1-peter/sshd.c Mon Mar 18 22:03:37 2002 *************** *** 264,273 **** main_sigchld_handler(int sig) { int save_errno = errno; ! int status; ! while (waitpid(-1, &status, WNOHANG) > 0) ! ; signal(SIGCHLD, main_sigchld_handler); errno = save_errno; --- 264,274 ---- main_sigchld_handler(int sig) { int save_errno = errno; ! int status, code; ! ! wh...

...& SSH_PROTO_1) + generate_ephemeral_server_key(); + + /* Arrange to restart on SIGHUP. The handler needs listen_sock. */ + signal(SIGHUP, sighup_handler); + + signal(SIGTERM, sigterm_handler); + signal(SIGQUIT, sigterm_handler); + + /* Arrange SIGCHLD to be caught. */ + signal(SIGCHLD, main_sigchld_handler); + + /* Write out the pid file after the sigterm handler is setup */ if (!debug_flag) { /* * Record our pid in /var/run/sshd.pid to make it @@ -871,17 +884,6 @@ fclose(f); } } - if (options.protocol & SSH_PROTO_1) - generate_ephemeral_server_key(); - - /* Arrange...

...6,10 @@ signal(SIGHUP, sighup_handler); signal(SIGTERM, sigterm_handler); signal(SIGQUIT, sigterm_handler); +#ifdef SIGDANGER + /* Don't die on AIX when the machine runs low on memory */ + signal(SIGDANGER, SIG_IGN); +#endif /* Arrange SIGCHLD to be caught. */ signal(SIGCHLD, main_sigchld_handler);

https://bugzilla.mindrot.org/show_bug.cgi?id=1692 Summary: sshd sometimes dies when sent multiple SIGHUPs in quick succession Product: Portable OpenSSH Version: 5.3p1 Platform: Other URL: https://bugs.launchpad.net/ubuntu/+source/openssh/+bug /497781 OS/Version: Linux Status: NEW

Hello! I'm hoping that Gmail won't HTML format this mail so that I'll get flamed :) Anyway, my question relates to ssh_config. The problem I find is that the Host pattern is only applied to the argument given on the command line, as outlined in the man page: "The host is the hostname argument given on the command line (i.e. the name is not converted to a canonicalized host name

...-z certificate_file]\n" +" [user@]hostname [command]\n" ); exit(255); } @@ -215,6 +216,7 @@ usage(void) static int ssh_session(void); static int ssh_session2(void); static void load_public_identity_files(void); +static void load_certificate_files(void); static void main_sigchld_handler(int); /* from muxclient.c */ @@ -595,7 +597,7 @@ main(int ac, char **av) again: while ((opt = getopt(ac, av, "1246ab:c:e:fgi:kl:m:no:p:qstvx" - "ACD:E:F:GI:KL:MNO:PQ:R:S:TVw:W:XYy")) != -1) { + "ACD:E:F:GI:KL:MNO:PQ:R:S:TVw:W:XYyz:")) != -1) { swi...

https://bugzilla.mindrot.org/show_bug.cgi?id=2646 Bug ID: 2646 Summary: zombie processes when using privilege separation Product: Portable OpenSSH Version: 7.2p2 Hardware: ix86 OS: Linux Status: NEW Severity: minor Priority: P5 Component: sshd Assignee: unassigned-bugs at

Numerous how-tos all over the Internet show how one would set up a tunnel using ssh, e.g.: ssh -f -o Tunnel=ethernet <server_ip> true I was wondering if there's a way to subsequently acquire the names of the local and remote tun/tap interfaces (e.g., using the default "-w any:any") for subsequent automatic tunnel configuration, e.g.: ip link set $TapDev up ip link set

In sshconnect.c there are two global variables for server_version_string client_version_string. These are used just in a few functions and can easily be passed as parameters. Also, there is a strange construct, where their memory is allocated to the global pointers, then copies of these pointers are assigned to the kex structure. The kex_free finally frees them via cleanup of the kex