Opened 2 years ago

Closed 2 years ago

Last modified 3 months ago

#64192 closed defect (fixed)

virt-manager @3.2.0_2 [quartz, [+]x11] Monterey python GTK crash

Reported by: GRMrGecko (Mr. Gecko) Owned by: mascguy (Christopher Nielsen)
Priority: Normal Milestone:
Component: ports Version: 2.7.1
Keywords: monterey arm64 Cc: michaelld (Michael Dickens)
Port: virt-manager

Description

Hello,

I'm trying to get virt-manager setup on my mac to manage remote hosts. I installed using the standard port install, without any modifications. Started dbus, and ran with the following command.

virt-manager --no-fork -c 'qemu+ssh://grmrgecko@10.0.0.5/system'

Upon running, it seems to crash with a GTK memory move error with a SIGTRAP. Google searching, I found ticket #63891 which looks to have a similar error occurring with another python GTK program. This means the error is likely at the python GTK level and not specifically this port. But I figured it would be worth opening a ticket to help others track this issue.

Attachments (1)

virt-manager-crash.txt (44.3 KB) - added by GRMrGecko (Mr. Gecko) 2 years ago.
Crash report from MacOS.

Download all attachments as: .zip

Change History (12)

Changed 2 years ago by GRMrGecko (Mr. Gecko)

Attachment: virt-manager-crash.txt added

Crash report from MacOS.

comment:1 Changed 2 years ago by mascguy (Christopher Nielsen)

Owner: set to mascguy
Status: newassigned

comment:2 Changed 2 years ago by mascguy (Christopher Nielsen)

Cc: michaelld added

It certainly could be related to issue:63891, but I'll need to take a closer look.

Also CCing @michaelld, for visibility and collaboration.

comment:3 Changed 2 years ago by michaelld (Michael Dickens)

Yes this looks like the same error. Traceback is identical for the last ~15 calls. I strongly doubt this issue is GNU Radio or Virt specific, but rather something to do with code in PyGObject or GTK3 or glib or (more generally) Gnome.

comment:4 Changed 2 years ago by michaelld (Michael Dickens)

It's actually good that that this one is with +x11 rather than +quartz ... I for years have used +quartz & thus often wonder when there is a bug whether that is the issue. Guessing not with this one, which reduces the search space for the bug .. yay!

comment:5 Changed 2 years ago by slarew

Cc: slarew added

comment:6 Changed 2 years ago by dm3jan (Janek)

are you on M1 or Intel?

comment:7 Changed 2 years ago by michaelld (Michael Dickens)

The crash report is for ARM64.

For folks with this issue and -not- building as +universal, can you please try the following:

Edit the glib2 Portfile. Find the entry for universal.patch and remove that line & save the file. Now build glib2 from source, uninstalling the prior version before installing this version. That takes care of this issue for me & should for you too. I think I've come to a conclusion as to how to fix the issue for glib2 2.64 ... but not yet for the current release (2.70).

comment:8 Changed 2 years ago by michaelld (Michael Dickens)

In 0519ce19a5703e6dc556d87a782059c53bcb7698/macports-ports (master):

glib2: fix section of universal.patch impacting ARM64 to match glib2-devel

Ref (should close, but let's get external feedback):
#64377
#64192
#63891

comment:9 Changed 2 years ago by mascguy (Christopher Nielsen)

@GRMrGecko, are you still seeing crashes, after updating all of your ports?

comment:10 Changed 2 years ago by mascguy (Christopher Nielsen)

Resolution: fixed
Status: assignedclosed

I believe this was resolved via the fixes for various ports, including @michaelld's changes for glib2.

If this is still an issue, please let us know!

comment:11 Changed 3 months ago by slarew

Cc: slarew removed
Note: See TracTickets for help on using tickets.