Bug 1273156
| Summary: | GDM does not work with XDMCP indirect | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | Red Hat Enterprise Linux 7 | Reporter: | Wolfgang Baudler <wbaudler> | ||||||||
| Component: | gdm | Assignee: | Ray Strode [halfline] <rstrode> | ||||||||
| Status: | CLOSED ERRATA | QA Contact: | Desktop QE <desktop-qa-list> | ||||||||
| Severity: | medium | Docs Contact: | |||||||||
| Priority: | unspecified | ||||||||||
| Version: | 7.1 | CC: | jkoten, mboisver, modehnal, rstrode, tpelka, wbaudler | ||||||||
| Target Milestone: | rc | ||||||||||
| Target Release: | --- | ||||||||||
| Hardware: | All | ||||||||||
| OS: | Linux | ||||||||||
| Whiteboard: | |||||||||||
| Fixed In Version: | gdm-3.22.3-6.el7 | Doc Type: | If docs needed, set a value | ||||||||
| Doc Text: | Story Points: | --- | |||||||||
| Clone Of: | Environment: | ||||||||||
| Last Closed: | 2017-08-01 22:57:55 UTC | Type: | Bug | ||||||||
| Regression: | --- | Mount Type: | --- | ||||||||
| Documentation: | --- | CRM: | |||||||||
| Verified Versions: | Category: | --- | |||||||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||||||
| Embargoed: | |||||||||||
| Attachments: |
|
||||||||||
|
Description
Wolfgang Baudler
2015-10-19 19:03:16 UTC
Xephyr -query <ip.address> should have read Xephyr -indirect <ip.address> what's the output of iptables -L (on both machines) Iptables is at the RHEL7.1 default on both machines Chain INPUT (policy ACCEPT) target prot opt source destination ACCEPT udp -- anywhere anywhere udp dpt:domain ACCEPT tcp -- anywhere anywhere tcp dpt:domain ACCEPT udp -- anywhere anywhere udp dpt:bootps ACCEPT tcp -- anywhere anywhere tcp dpt:bootps Chain FORWARD (policy ACCEPT) target prot opt source destination ACCEPT all -- anywhere 192.168.122.0/24 ctstate RELATED,ESTABLISHED ACCEPT all -- 192.168.122.0/24 anywhere ACCEPT all -- anywhere anywhere REJECT all -- anywhere anywhere reject-with icmp-port-unreachable REJECT all -- anywhere anywhere reject-with icmp-port-unreachable Chain OUTPUT (policy ACCEPT) target prot opt source destination ACCEPT udp -- anywhere anywhere udp dpt:bootpc Even after a "iptables --flush" on both machines, I still get the same result. Any update on this problem? It has been almost a year now. can you add Enable=true to the [debug] section of /etc/gdm/custom.conf, reboot, reproduce and then attach /var/log/messages and /var/log/gdm/* ? Created attachment 1202988 [details]
/var/log/gdm/:0.log
Enable=true in the [debug] section was already set. Re-tested with RHEL 7.2 and the same problem still exists (black screen instead of usable XDMCP chooser). /var/log/messages: Sep 20 13:36:48 rhel7-test gdm: GLib-GObject: g_object_new: assertion 'G_TYPE_IS_OBJECT (object_type)' failed Sep 20 13:36:48 rhel7-test gdm: GLib-GObject: invalid (NULL) pointer instance Sep 20 13:36:48 rhel7-test gdm: GLib-GObject: g_signal_connect_object: assertion 'G_TYPE_CHECK_INSTANCE (instance)' failed Sep 20 13:36:48 rhel7-test gdm: GLib-GObject: invalid (NULL) pointer instance Sep 20 13:36:48 rhel7-test gdm: GLib-GObject: g_signal_connect_object: assertion 'G_TYPE_CHECK_INSTANCE (instance)' failed Sep 20 13:36:48 rhel7-test gdm: GLib-GObject: g_object_bind_property_full: assertion 'G_IS_OBJECT (source)' failed Sep 20 13:36:48 rhel7-test gdm: GLib-GObject: invalid (NULL) pointer instance Sep 20 13:36:48 rhel7-test gdm: GLib-GObject: g_signal_connect_data: assertion 'G_TYPE_CHECK_INSTANCE (instance)' failed Sep 20 13:36:48 rhel7-test gdm: gdm_slave_start: assertion 'GDM_IS_SLAVE (slave)' failed Sep 20 13:36:50 rhel7-test gdm: gdm_slave_start: assertion 'GDM_IS_SLAVE (slave)' failed Sep 20 13:36:54 rhel7-test gdm: gdm_slave_start: assertion 'GDM_IS_SLAVE (slave)' failed The GDM_IS_SLAVE failed messages repeat about every 16 seconds. /var/log/gdm/* files added as attachments. Created attachment 1202989 [details]
/var/log/gdm/:0-greeter.log
can you attach /var/log/messages ? Created attachment 1203004 [details]
/var/log/messages showing gdm startup and XDMCP connection attempt
I believe the problem is this code:
g_object_class_install_property (object_class,•
PROP_SLAVE_TYPE,•
g_param_spec_gtype ("slave-type",•
"slave type",•
"slave type",•
GDM_TYPE_SIMPLE_SLAVE,•
G_PARAM_READWRITE | G_PARAM_CONSTRUCT));•
inside gdm-display.c.
The GDM_TYPE_SIMPLE_SLAVE there is too specific. It should just say the more general GDM_TYPE_SLAVE, so that a slave type of GDM_TYPE_XDMCP_CHOOSER_SLAVE is allowed, as specified in gdm-xdmcp-chooser-display.c.
This bug was introduced as part of the cleanup work done upstream here:
https://bugzilla.gnome.org/show_bug.cgi?id=726380
I have two 7.4 machines set up, one with the custom.conf specified and one I am using to try to connect to it. Both have gdm-3.22.3-11. I start both in runlevel 3 and run "X -indirect <ip address> -nolisten inet6" I only get a black screen still. Am I missing something or can you give it a try with your setup Wolfgang? I don't have a RHEL 7.4 (beta?) machine at the moment, but I just now re-tested this on RHEL 7.3 with gdm-3.14.2-20.el7_3.x86_64 and the problem still persists. All I get is the black screen. I will try to get a RHEL7.4 test setup, but your report with gdm-3.22.3-11 is not encouraging. It has been quite a while since I reported this. What needs to happen to get this bug fixed? It is preventing upgrades to RHEL7 on our XDMCP servers. (In reply to Wolfgang Baudler from comment #22) > I don't have a RHEL 7.4 (beta?) machine at the moment, but I just now > re-tested this on RHEL 7.3 with gdm-3.14.2-20.el7_3.x86_64 and the problem > still persists. All I get is the black screen. > > I will try to get a RHEL7.4 test setup, but your report with gdm-3.22.3-11 > is not encouraging. > > It has been quite a while since I reported this. What needs to happen to get > this bug fixed? It is preventing upgrades to RHEL7 on our XDMCP servers. I am able to get Xephyr to work correctly and as you described. Can you elaborate on the exact setup in order for the "X -indirect <ip address> -nolisten inet6" command to previously function? X or Xephyr should not matter. Using Xephyr is just easier to use for testing this. Xephyr works for you? You get a XDMCP chooser window displayed as it should when connecting to a RHEL7 gdm display manager? This works for me only with RHEL6, which uses kdm, not gdm. I have actually not tried with gdm on RHEL6. But since kdm is no longer available on RHEL7, it is necessary to make it work with gdm now. I stand corrected. I cannot connect a 7.4 to a 7.4 machine using either of the methods. I am able to use Xephyr to connect to the 7.4 "host" from a Fedora 24 "client." Both methods provide a black screen with using 7.4 machines. Also note, multiple restarts and iptables -F didn't make a difference. gdm-3.22.3-11.el7 on both machines. Hi Wolfgang, Mike is on the QE team and is testing some fixes that are designed to address this bug report for the next synchronous update of Red Hat Enterprise Linux (7.4). Based on testing feedback from Mike, those fixes are either incomplete or there's a configuration problem with Mike's setup. Red Hat will continue to investigate and report back. Ray came over and found an issue with the firewall on my "client" machine. Fortunately it was a setup issue and I am now able to use both methods described in this bug to see the XDMCP greeter and subsequently log into an XDMCP session. With RHEL 7.4 on both client and server machines? (In reply to Wolfgang Baudler from comment #28) > With RHEL 7.4 on both client and server machines? Correct, both machines have RHEL 7.4 Workstation installed on them. Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2017:2128 |