wineboot hangs when running on a domain login

[acsheehy]

Whenever I create a new prefix, (not running as root, nor sudo'd - but as a standard domain user on a domain-joined PC) by any method (plain wine cli, but also qt4wine and winetricks gives the same result!) wineboot hangs while updating the prefix.

0014:err:ole:marshal_object couldn't get IPSFactory buffer for interface {00000131-0000-0000-c000-000000000046}
0014:err:ole:marshal_object couldn't get IPSFactory buffer for interface {6d5140c1-7436-11ce-8034-00aa006009fa}
0014:err:ole:StdMarshalImpl_MarshalInterface Failed to create ifstub, hres=0x80004002
0014:err:ole:CoMarshalInterface Failed to marshal the interface {6d5140c1-7436-11ce-8034-00aa006009fa}, 80004002
0014:err:ole:get_local_server_stream Failed: 80004002
0012:err:ole:marshal_object couldn't get IPSFactory buffer for interface {00000131-0000-0000-c000-000000000046}
0012:err:ole:marshal_object couldn't get IPSFactory buffer for interface {6d5140c1-7436-11ce-8034-00aa006009fa}
0012:err:ole:StdMarshalImpl_MarshalInterface Failed to create ifstub, hres=0x80004002
0012:err:ole:CoMarshalInterface Failed to marshal the interface {6d5140c1-7436-11ce-8034-00aa006009fa}, 80004002
0012:err:ole:get_local_server_stream Failed: 80004002

followed by, eventually:
0009:err:process:__wine_kernel_init boot event wait timed out

It appears that wine cannot find something, maybe??? Reinstalled, different versions, all have the same result. I have searched high and low for a couple of weeks now, but cannot find anything to help. Anyone have any ideas?

A

Debian 9
wine 3.0 stable (but I have tried older and staging versions)
pbis-open 8.5.385.1
Domain Login on a Domain PC
Dell Optiplex 9020

[Bob Wya]

@acsheehy

Try launching your wineserver process separately first (for the WINEPREFIX you are testing):

Code: Select all

wineserver -d2 -f

Maybe that log will indicate why the wineserver process is hanging.

Please use the forum Code:

Code: Select all

Code

tags for terminal output and commands.

Ta
Bob

[acsheehy]

The result is

Code: Select all

me@pc:~$ WINEPREFIX=~/.passgen wineserver -d2 -f
sock_init: shutdown() causes EOF
wineserver: starting (pid=18493)
Not saving clean key \\Machine
Not saving clean key \\User\\.Default
Not saving clean key \\User\\S-1-5-21-0-0-0-1000
Not saving clean key \\Machine
Not saving clean key \\User\\.Default
Not saving clean key \\User\\S-1-5-21-0-0-0-1000
Not saving clean key \\Machine
Not saving clean key \\User\\.Default
Not saving clean key \\User\\S-1-5-21-0-0-0-1000

Ad nauseum. Thanks for the pointer...

[Bob Wya]

@acsheehy,

That looks OK.

So you can further verify that the wineserver process is setting up a listening socket (while the above process is running):

Code: Select all

lsof -U -T -R | grep 18493 # <--- insert wineserver PID here

You should also "see" the socket in your tmp filesystem, e.g.

Code: Select all

ls -l /tmp/.wine-1000/server-811-2960024/
total 0
-rw------- 1 robert users 0 Jan 31 16:58 lock
srw------- 1 robert users 0 Jan 31 16:58 socket

Interestingly your issue appear to mirror this WineHQ Bug Report: Bug 43784 - Updating a 2.16 prefix to 2.17 causes Wine to hang forever updating the configuration.

I think it would be safe to "hijack" that bug report (your Wine issues are pretty similar) and put a log attachment with:

Code: Select all

export WINEDEBUG=+tid,+timestamp,+all
wineboot -u &>~/wine_boot_log.txt

If that log goes over 5000Kbytes then please compress it with xz (xz -9 -e ).
Then attach the log to the bug report - mentioning it affects the Wine stable 3.0 release.

I'll subscribe to the bug report and poke about in the log file - when you post it.

Bob

[acsheehy]

Yes, you are right:

lsof produces:

Code: Select all

wineserve 11195  2903 anthony.sheehy    3u  unix 0xffffa0fc90b1b800      0t0 828501 type=STREAM
wineserve 11195  2903 anthony.sheehy    9u  unix 0xffffa0fc395e9c00      0t0 828502 socket type=STREAM
wineserve 11195  2903 anthony.sheehy   19u  unix 0xffffa0fb8e6cdc00      0t0 825170 socket type=STREAM
wineserve 11195  2903 anthony.sheehy   24u  unix 0xffffa0fc90871c00      0t0 828513 type=STREAM
wineserve 11195  2903 anthony.sheehy   28u  unix 0xffffa0fc395e9400      0t0 828516 type=STREAM
wineserve 11195  2903 anthony.sheehy   37u  unix 0xffffa0fc3dd7c400      0t0 827927 type=STREAM
wineserve 11195  2903 anthony.sheehy   42u  unix 0xffffa0fbb88dc000      0t0 828521 type=STREAM
wineserve 11195  2903 anthony.sheehy   48u  unix 0xffffa0fc9332fc00      0t0 826969 type=STREAM
wineserve 11195  2903 anthony.sheehy   50u  unix 0xffffa0fc8fd16800      0t0 827974 type=STREAM
wineserve 11195  2903 anthony.sheehy   53u  unix 0xffffa0fc9332f800      0t0 827941 type=STREAM
wineserve 11195  2903 anthony.sheehy   57u  unix 0xffffa0fba89a1c00      0t0 825217 type=STREAM
wineserve 11195  2903 anthony.sheehy   58u  unix 0xffffa0fb8e7f3400      0t0 828535 type=STREAM

and the temp files:

total 0
-rw------- 1 anthony.sheehy domain^users 0 Feb 7 13:36 lock
srw------- 1 anthony.sheehy domain^users 0 Feb 7 13:36 socket

And the debug text file just keeps on filling up...

[acsheehy]

There we go - did I get it right?

[Bob Wya]

There we go - did I get it right?

Looking good... Well if you like challenges anyway!

I had a cursory look at the attachment on the bug report...
The issue looks, at first glance, like one of those horrible timing deadlock thingy-me-bobs!

Anyway - hopefully with a detailed log like that - someone more knowledgable than me can pick it up (hopefully)!
I'll try and schedule in some more poking about with the log myself as well.

Thanks
Bob