Bug 1938168
| Summary: | GUI dies with backtrace "TypeError: unknown type (null)" | ||||||
|---|---|---|---|---|---|---|---|
| Product: | Red Hat Enterprise Linux 8 | Reporter: | Renaud Métrich <rmetrich> | ||||
| Component: | anaconda | Assignee: | Radek Vykydal <rvykydal> | ||||
| Status: | CLOSED ERRATA | QA Contact: | Release Test Team <release-test-team-automation> | ||||
| Severity: | urgent | Docs Contact: | |||||
| Priority: | urgent | ||||||
| Version: | 8.3 | CC: | anaconda-maint-list, jkonecny, jstodola, lmiksik, pgm-rhel-tools, pratshar, pzatko, rvykydal, sbarcomb, thashimo, zveleba | ||||
| Target Milestone: | rc | Keywords: | Triaged | ||||
| Target Release: | --- | ||||||
| Hardware: | x86_64 | ||||||
| OS: | Linux | ||||||
| Whiteboard: | |||||||
| Fixed In Version: | anaconda-33.16.4.14-1.el8 | Doc Type: | Bug Fix | ||||
| Doc Text: |
Cause:
If network was activated shortly before visiting Time & Date spoke and opening of the dialog for NTP servers configuration in the spoke, Anaconda performed NTP status check on an invalid internal object.
f.
Consequence:
Installation crashed with a traceback.
Fix:
The check of object created when entering the Time & Date spoke and invalidated by opening the dialog for NTP server was made more robust.
Result:
Anaconda no more crashes in case of opening of NTP servers configuration dialog shortly after network activation.
|
Story Points: | --- | ||||
| Clone Of: | |||||||
| : | 1940472 (view as bug list) | Environment: | |||||
| Last Closed: | 2021-05-18 15:47:16 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: | |||||||
| Bug Depends On: | |||||||
| Bug Blocks: | 1940472, 1944724 | ||||||
| Attachments: |
|
||||||
Also see bug 1917797, which has the same traceback and a reproducer that was confirmed to be working. Thank you for the reproducer Jan, setting dev_ack+. Updates image for RHEL 8.3 containing the fix from comment #18: https://rvykydal.fedorapeople.org/updates.rhbz1938168.img It is possible that fetching the image from network using boot option inst.updates=https://rvykydal.fedorapeople.org/updates.rhbz1938168.img can spoil the reproducer (configuring network in early stage / initramfs) so using hd: or cdrom: variants of inst.updates may be needed: https://anaconda-installer.readthedocs.io/en/latest/boot-options.html#inst-updates Hi Radek, With the update image, I can still reproduce the crash on the Subscription Manager spoke. I guess only the NTP panel is fixed here. Could you please attach the logs from the crash on the Subscription Manager spoke ? Created attachment 1765209 [details]
Crash in Subscription Manager spoke when proxy is badly configured
Hi Renaud, the traceback from comment 23 is a different issue than the NTP problem originally reported in this bug. Can you please create a new bug for it? Please, also describe the reproducer there. This bug will track only the initial problem with NTP. Thank you for understanding. OK I filed BZ #1941578 *** Bug 1917797 has been marked as a duplicate of this bug. *** Switching to Verified - fix is present in RHEL-8.4.0-20210324.n.0 I've updated the Doc Text field. Adding more detailed info here in a comment, hope it helps to understand the cause of the issue: There are two checks of the list of NTP servers, each run in a separate thread (ie concurrently): 1) one when entering the spoke 2) another one when entering NTP servers configuration dialog in the spoke The 2) invalidates some objects checked in 1) (it clears the list used in 1)) so if it happens before 1) is finished there is the crash. And the condition in which the check 1) takes so long that it collides with the check initiated by 2) is that network is activated only shortly before the NTP configuration (ie not at the early stage of installation) or takes too long. (More technical explanation is in the PR https://github.com/rhinstaller/anaconda/pull/3252 for the fix.) *** Bug 1940472 has been marked as a duplicate of this bug. *** 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 (anaconda bug fix and enhancement update), 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/RHBA-2021:1844 |
Description of problem: A customer trying to install RHV4.4-20210202.0 (based on RHEL8.3) cannot install using the graphical interface because he gets the following backtrace shortly after configuring its NTP servers: -------- 8< ---------------- 8< ---------------- 8< ---------------- 8< -------- The following was filed automatically by anaconda: anaconda 33.16.3.26 exception report Traceback (most recent call first): File "/usr/lib64/python3.6/site-packages/gi/overrides/Gtk.py", line 1099, in __getitem__ return self.model.get_value(self.iter, key) File "/usr/lib64/python3.6/site-packages/pyanaconda/ui/gui/spokes/datetime_spoke.py", line 326, in _set_server_ok_nok actual_hostname = self._serversStore[itr][SERVER_HOSTNAME] File "/usr/lib64/python3.6/threading.py", line 864, in run self._target(*self._args, **self._kwargs) File "/usr/lib64/python3.6/site-packages/pyanaconda/threading.py", line 280, in run threading.Thread.run(self) TypeError: unknown type (null) -------- 8< ---------------- 8< ---------------- 8< ---------------- 8< -------- Version-Release number of selected component (if applicable): 33.16.3.26-1 How reproducible: Always on customer system, didn't manager on my lab systems Steps to Reproduce: