Bug 1883868

Summary:	[RFE] Dump JS stack trace by default when gnome-shell crashes
Product:	Red Hat Enterprise Linux 8	Reporter:	Phil Jasbutis <pjasbuti>
Component:	gnome-shell	Assignee:	Florian Müllner <fmuellner>
Status:	CLOSED ERRATA	QA Contact:	Tomas Pelka <tpelka>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	8.2	CC:	jadahl, sbarcomb, tpelka
Target Milestone:	rc	Keywords:	FutureFeature, Triaged
Target Release:	8.0
Hardware:	All
OS:	Linux
Whiteboard:
Fixed In Version:	gnome-shell-3.32.2-27.el8	Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2021-05-18 14:38:50 UTC	Type:	Feature Request
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Phil Jasbutis 2020-09-30 12:37:48 UTC

3. What is the nature and description of the request?

- after facing issues+crashes with gnome-shell it turned out that debugging options are quite limited at the moment
- customer tested DEBUG_SHELL="backtrace-segfaults" in .bash_profile to get a JS stack trace when gnome-shell crashed
- customer would like to have this implemented as a default for every gnome-shell which runs on a host

4. Why does the customer need this? (List the business requirements here)

- to better understand gnome-shell crashes in retrospect and support debugging gnome-shell issues
- if gnome-shell crashed and re-spawned there is no indicator in the logs by default

5. How would the customer like to achieve this? (List the functional requirements here)

- set environment variable DEBUG_SHELL="backtrace-segfaults" as default in gnome-shell so every gnome-shell process on this
host (independent from the user) dumps a JS stack trace in journal when gnome-shell crashes.

6. For each functional requirement listed, specify how Red Hat and the customer can test to confirm the requirement is successfully implemented.

- start a gnome-shell in userspace, open windows and gnome applications
- force the gnome-shell to crash by sending SEGV to the process
- check if the journal contains at least the header for the stack trace when crashed
- e.g. "Sep 02 12:30:43 <host> org.gnome.Shell.desktop[7197]: == Stack trace for context 0x26820c0 =="

7. Is there already an existing RFE upstream or in Red Hat Bugzilla?

- not known

8. Does the customer have any specific timeline dependencies and which release would they like to target (i.e. RHEL5, RHEL6)?

- no, asap

9. Is the sales team involved in this request and do they have any additional input?

- no

10. List any affected packages or components.

- gnome-shell

11. Would the customer be able to assist in testing this functionality if implemented?

yes

Comment 6 Florian Müllner 2020-10-29 17:31:41 UTC

(In reply to Phil Jasbutis from comment #0)
> > - if gnome-shell crashed and re-spawned there is no indicator in the logs by
> default

That's not true, there should be a line along the lines of

  Oct 29 18:06:31 rhel7-unknow gnome-session[1532]: gnome-session-binary[1532]: WARNING: Application 'org.gnome.Shell.desktop' killed by signal 11


> - set environment variable DEBUG_SHELL="backtrace-segfaults" as default in
> gnome-shell so every gnome-shell process on this
> host (independent from the user) dumps a JS stack trace in journal when
> gnome-shell crashes.

That said, treating an unset DEBUG_SHELL variable as "backtrace-segfaults" is certainly feasible.

Comment 14 errata-xmlrpc 2021-05-18 14:38:50 UTC

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 (Moderate: GNOME security, 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/RHSA-2021:1586