Hi, I see in the release notes a reference to a delay when logging on (or su¹ing). I assume you know the cause, in case an strace shows a timeout when polling on some dbus communication:
36237 12:44:18.552016 connect(3, {sa_family=AF_FILE, path="/var/run/dbus/system_bus_socket"}, 33) = 0 36237 12:44:18.554508 fcntl(3, F_GETFL) = 0x2 (flags O_RDWR) 36237 12:44:18.555690 fcntl(3, F_SETFL, O_RDWR|O_NONBLOCK) = 0 36237 12:44:18.557008 geteuid() = 0 36237 12:44:18.558264 getsockname(3, {sa_family=AF_FILE, NULL}, [2]) = 0 36237 12:44:18.562788 poll([{fd=3, events=POLLOUT}], 1, 0) = 1 ([{fd=3, revents=POLLOUT}]) 36237 12:44:18.564702 send(3, "\0", 1, MSG_NOSIGNAL) = 1 36237 12:44:18.568214 send(3, "AUTH EXTERNAL 30\r\n", 18, MSG_NOSIGNAL) = 18 36237 12:44:18.570001 poll([{fd=3, events=POLLIN}], 1, -1) = 1 ([{fd=3, revents=POLLIN}]) 36237 12:44:18.573616 read(3, "OK 30473130807de44448e948ea00000"..., 2048) = 37 36237 12:44:18.575757 poll([{fd=3, events=POLLOUT}], 1, -1) = 1 ([{fd=3, revents=POLLOUT}]) 36237 12:44:18.577083 send(3, "NEGOTIATE_UNIX_FD\r\n", 19, MSG_NOSIGNAL) = 19 36237 12:44:18.581047 poll([{fd=3, events=POLLIN}], 1, -1) = 1 ([{fd=3, revents=POLLIN}]) 36237 12:44:18.582391 read(3, "AGREE_UNIX_FD\r\n", 2048) = 15 36237 12:44:18.583652 poll([{fd=3, events=POLLOUT}], 1, -1) = 1 ([{fd=3, revents=POLLOUT}]) 36237 12:44:18.585250 send(3, "BEGIN\r\n", 7, MSG_NOSIGNAL) = 7 36237 12:44:18.589419 poll([{fd=3, events=POLLIN|POLLOUT}], 1, -1) = 1 ([{fd=3, revents=POLLOUT}]) 36237 12:44:18.590870 sendmsg(3, {msg_name(0)=NULL, msg_iov(2)=[{"B\1\0\1\0\0\0\0\0\0\0\1\0\0\0n\1\1o\0\0\0\0\25/org/fre"..., 128}, {"", 0}], msg_controllen=0, msg_flags=0}, MSG_NOSIGNAL) = 128 36237 12:44:18.612217 poll([{fd=3, events=POLLIN}], 1, 25000) = 1 ([{fd=3, revents=POLLIN}]) 36237 12:44:18.614895 recvmsg(3, {msg_name(0)=NULL, msg_iov(1)=[{"B\2\1\1\0\0\0\n\0\0\0\1\0\0\0=\6\1s\0\0\0\0\5:1.60\0\0\0"..., 2048}], msg_controllen=0, msg_flags=MSG_CMSG_CLOEXEC}, MSG_CMSG_CLOEXEC) = 260 36237 12:44:18.624871 recvmsg(3, 0x3ffffd901d0, MSG_CMSG_CLOEXEC) = -1 EAGAIN (Resource temporarily unavailable) 36237 12:44:18.645198 sendmsg(3, {msg_name(0)=NULL, msg_iov(2)=[{"B\1\0\1\0\0\0\\0\0\0\2\0\0\0\226\1\1o\0\0\0\0\27/org/fre"..., 168}, {"\0\0\0\0\0\0\215\215\0\0\0\2su\0\0\0\0\0\3tty\0\0\0\0\0\0\0\0\0"..., 92}], msg_controllen=0, msg_flags=0}, MSG_NOSIGNAL) = 260 36237 12:44:18.648527 poll([{fd=3, events=POLLIN}], 1, 25000) = 0 (Timeout) 36237 12:44:43.684151 open("/etc/localtime", O_RDONLY) = 4
Neale
Neale Ferguson píše v St 21. 12. 2011 v 11:49 -0600:
Hi, I see in the release notes a reference to a delay when logging on (or su¹ing). I assume you know the cause, in case an strace shows a timeout when polling on some dbus communication:
thanks for the info, I suspect that systemd doesn't start something or it fails or something like that. It's definitely something I'll look at.
Dan
Further:
Dec 19 18:47:48 fedora16 dbus[1097]: [system] Activating via systemd: service name='org.freedesktop.login1' unit='dbus-org.freedesktop.loginervice' Dec 19 18:47:48 fedora16 dbus-daemon[1097]: dbus[1097]: [system] Activating via systemd: service name='org.freedesktop.login1' unit='dbus-orreedesktop.login1.service' Dec 19 18:47:48 fedora16 systemd-logind[1203]: Failed to open /sys/class/tty/tty0/active: No such file or directory Dec 19 18:47:48 fedora16 systemd-logind[1203]: Failed to fully start up daemon: No such file or directory Dec 19 18:47:48 fedora16 systemd[1]: systemd-logind.service: main process exited, code=exited, status=1 Dec 19 18:47:49 fedora16 systemd[1]: Unit systemd-logind.service entered failed state.
I'm trying to track things further. I'll let you know how I get on.
On 12/22/11 3:48 AM, "Dan Horák" dan@danny.cz wrote:
Neale Ferguson píše v St 21. 12. 2011 v 11:49 -0600: Hi, I see in the release notes a reference to a delay when logging on (or su¹ing). I assume you know the cause, in case an strace shows a timeout when polling on some dbus communication:
thanks for the info, I suspect that systemd doesn't start
something or
it fails or something like that. It's definitely something I'll
look at.
Dan
_______________________________________________ s390x mailing
list
s390x@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/lis
tinfo/s390x
Problem also occurs in openSUSE:
http://lists.opensuse.org/opensuse-bugs/2011-12/msg00945.html
"There is no tty0 on zLinux, only the console device:"
Patch from the systemd-devel list:
diff --git a/src/logind.c b/src/logind.c index 4633a5e..f491cee 100644 --- a/src/logind.c +++ b/src/logind.c @@ -906,6 +906,8 @@ static int manager_connect_console(Manager *m) {
m->console_active_fd = open("/sys/class/tty/tty0/active", O_RDONLY|O_NOCTTY|O_CLOEXEC); if (m->console_active_fd < 0) { + if (errno == ENOENT) + return 0; log_error("Failed to open /sys/class/tty/tty0/active: %m"); return -errno; }
Neale Ferguson píše v Čt 22. 12. 2011 v 09:29 -0600:
Problem also occurs in openSUSE:
http://lists.opensuse.org/opensuse-bugs/2011-12/msg00945.html
"There is no tty0 on zLinux, only the console device:"
Patch from the systemd-devel list:
it's fixed in systemd master branch for a while and I've opened a bug [1] in Fedora to get an updated systemd package. Neale, thanks again for your investigation of the issue.
Dan
Dan Horák píše v St 04. 01. 2012 v 19:32 +0100:
Neale Ferguson píše v Čt 22. 12. 2011 v 09:29 -0600:
Problem also occurs in openSUSE:
http://lists.opensuse.org/opensuse-bugs/2011-12/msg00945.html
"There is no tty0 on zLinux, only the console device:"
Patch from the systemd-devel list:
it's fixed in systemd master branch for a while and I've opened a bug [1] in Fedora to get an updated systemd package. Neale, thanks again for your investigation of the issue.
I've made a new build of systemd which should fix the login delay issue, download rpms directly from http://s390.koji.fedoraproject.org/koji/buildinfo?buildID=89914 until it's published as a regular update.
Dan
Dan
[1] https://bugzilla.redhat.com/show_bug.cgi?id=771579
s390x mailing list s390x@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/s390x
Dan Wrote
I've made a new build of systemd which should fix the login delay issue, download rpms directly from http://s390.koji.fedoraproject.org/koji/buildinfo?buildID=89914 until it's published as a regular update.
Dan,
That fix worked - the initial sign-on delay is gone now on FC16....
Thanks.... -Phil
-----Original Message----- From: s390x-bounces@lists.fedoraproject.org [mailto:s390x- bounces@lists.fedoraproject.org] On Behalf Of Dan Horák Sent: Wednesday, January 18, 2012 3:43 AM To: s390x@lists.fedoraproject.org Subject: Re: Login delay
Dan Horák píše v St 04. 01. 2012 v 19:32 +0100:
Neale Ferguson píše v Čt 22. 12. 2011 v 09:29 -0600:
Problem also occurs in openSUSE:
http://lists.opensuse.org/opensuse-bugs/2011-12/msg00945.html
"There is no tty0 on zLinux, only the console device:"
Patch from the systemd-devel list:
it's fixed in systemd master branch for a while and I've opened a bug [1] in Fedora to get an updated systemd package. Neale, thanks again for your investigation of the issue.
I've made a new build of systemd which should fix the login delay issue, download rpms directly from http://s390.koji.fedoraproject.org/koji/buildinfo?buildID=89914 until it's published as a regular update.
Dan
Dan
[1] https://bugzilla.redhat.com/show_bug.cgi?id=771579
s390x mailing list s390x@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/s390x
s390x mailing list s390x@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/s390x
This message has been scanned for viruses and dangerous content.
--------------------------------------------- This message has been scanned for viruses and dangerous content.