VMware Communities
lpatters35254
Contributor
Contributor

VMware Workstation-10.0.1 fails after RedHat-6 software update.

After updating software on a RedHat-6 system (yum update), workstation-10.0.1 will 'crash' the system.

The update included:

kernel-2.6.32-431

glib2-2.12-1.132

The 'Bug Tracking' tool reports problems with glib2 and kernel 'hang'.

Suggestion?

Thanks!

Tags (1)
83 Replies
kuespert
Enthusiast
Enthusiast

Thank you for your reply, but

> service vmware start

starts /etc/init.d/vmware start, nothing else.

My output of

> chkconfig --list | grep vmware

is different from yours. This is not surprising, because I have no workstation but a VMplayer:

Aus = off; Ein = on;

vmware                         0:Aus   1:Aus   2:Ein   3:Ein   4:Aus   5:Ein   6:Aus
vmware-USBArbitrator    0:Aus   1:Aus   2:Ein   3:Ein   4:Aus   5:Ein   6:Aus

no vmware-workstation-server

Greetings!

kuespert

Reply
0 Kudos
NobuoSha
Contributor
Contributor

The "service vmware start" basically check to see if the kernel modules are built for the current kernel or not. If not, it will build them and then load them. Once the kernel modules are built and loaded, it won't do anything. But the "service vmware status" will show:

Module vmmon loaded
Module vmnet loaded

If yours does not show this, then probably the kernel modules were not built properly (thus could not be loaded). Then it's a good idea to redo it:

service vmware stop

rm -v /lib/modules/$(uname -r)/misc/v*.ko

depmod -a

umask 022

service vmware start

Reply
0 Kudos
NobuoSha
Contributor
Contributor

Some comments about my last reply:

  1. These procedures are under the assumption that you have already done something like "mv -v /usr/lib/vmware/modules/binary /usr/lib/vmware/modules/binary.orig" before (and you indicated that, too.).
  2. The first step "service vmware stop" might not be necessary but shouldn't hurt.
  3. The "umask 022" is to make sure that the generated kernel modules are world readable.

Good luck!

Reply
0 Kudos
kuespert
Enthusiast
Enthusiast

After the procedure, I reported in post 57, /lib/modules/$(uname -r)/misc was empty, even though I performed depmod -a. If service vmware start tries to start modules from there, it finds nothing. depmod -a did not write anything to stdout or stderr.

Same behavier during a further try.

Greetings!

Reply
0 Kudos
NobuoSha
Contributor
Contributor

Does the /usr/lib/vmware/modules/source/ exist? If so, does it contain some tar files there? Probably you need to reinstall the VMware Player again and then repeat the procedures Jack10z posted. It works for many people including me.

Reply
0 Kudos
kuespert
Enthusiast
Enthusiast

Content of /usr/lib/vmware/modules/source

-rw-r--r--    1 root     root       757760 Oct 27 09:32 vmblock.tar
-rw-r--r--    1 root     root      1249280 Oct 27 09:32 vmci.tar
-rw-r--r--    1 root     root      1208320 Oct 27 09:32 vmmon.tar
-rw-r--r--    1 root     root       880640 Oct 27 09:32 vmnet.tar
-rw-r--r--    1 root     root      1024000 Oct 27 09:32 vsock.tar

Is it ok?

Reply
0 Kudos
NobuoSha
Contributor
Contributor

That looks right. Then I don't know why the kernel modules were not built. I will do more investigations when I have time. (Just to be sure that you did the 'service vmware start' as root, right?)

Reply
0 Kudos
okurth1
VMware Employee
VMware Employee

Hi kuespert,

can you please just start 'vmplayer' in a terminal window, and post the output?

If you don't have the modules compiled yet, there should be a dialog box popping up that informs you that modules will be built, but maybe it is missing something.

Reply
0 Kudos
kuespert
Enthusiast
Enthusiast

Hi NobuoSha and okurth1,

thank you very much for your help!

I started vmplayer and confirmed to the module generation. After the usual progress display the player Window appeared. (The VM was forgotten!?) The log file is attached. gsettings was aborted (report is attached).

Unfortunately it wrote more to the stderr than the xterm buffer could store. In /lib/modules/2.6.32-431.3.1.el6.x86_64/misc 5 v*.ko files were found.

In order to catch the stderr into a file, I removed them and started vmplayer again. stderr file is attached. Same behavior, but gsettings did not crash. After upload of this infos, I will try to start the VM.

Greetings!

Reply
0 Kudos
kuespert
Enthusiast
Enthusiast

Hi,

within my former post was a minor error: during the second run of vmplayer gsettings was aborted, too.

My next attempt was a start of vmplayer with VM argument: no message on stderr, but abort of gsettings; the most important result: VM runs correctely without problems. This is reproducible!

Then I updated the kernel to 2.6.32-431.5.1.el6.x86_64 and started vmplayer. During the modules update many error messages (see attachment!), gsettings is aborted, but the VM runs perfectly!

Summary:

- The main problem is solved: The VM may be driven.

- Unpleasant remain: error messages during the modules update and

                               the gsettings abort with abrt window and report mail.

PS (21.3.14): Since glib2 update 2.26.1-7.el6_5 no message about a gsettings abort!

Reply
0 Kudos
dkt
Contributor
Contributor

Hi NobuoSha,


Yes, it worked, in my Fedora 19.

Thanks.

Reply
0 Kudos
Summer88
Contributor
Contributor

Same situation with a Debian 7 host and a Win7 64bit guest, tried to get the tools working without success.

I've followed the procedure installed the package and recompiled the wmware.

When the guest start, it stuck at user login screen, after you've entered the password. It freezes.

Reply
0 Kudos
1stefan
Contributor
Contributor

Vielen Dank für Ihre E-Mail, ich bin derzeit leider krank.

Ich werde versuchen, Ihre E-Mail so schnell wie möglich zu beantworten.

Dear Sender, thank you for your e-mail. Unfortunately, I am ill.

In any case, I will reply to your mail as soon as possible.

Mit freundlichen Grüßen and with best regards,

Stefan Brozinski

Reply
0 Kudos
plo123
Contributor
Contributor

I can't offer a solution, but I can offer a workaround.

Uninstalling the vmware tools package from the Windows 7 guest appears to have resolved this issue for our users.  (or a similar issue with the same symptoms -- host crash on centos 6.5 win7 vm guest bootup)

annoying as vmware tools gives a lot of functionality, but at least our new 6.x users can use their VM's again.

Note that we are running CentOS 6.5 with the latest avaliable distro kernel as of today.

Reply
0 Kudos
okurth1
VMware Employee
VMware Employee

Summer88,

can you please tell the kernel version (output of "uname -a") ? We can then try to reproduce it here. Also, it's not clear to me what you mean by "it" freezes. Does the host freeze, or the guest?

Thanks,

Oliver

Reply
0 Kudos
Summer88
Contributor
Contributor

the kernel version is:

Linux server 3.12-1-amd64 #1 SMP Debian 3.12.9-1 (2014-02-01) x86_64 GNU/Linux

The guest OS, the Win7 freezes, I mean it get to login, you digit the password, the window change and stay on the light blue screen with the blue circle and the word Windows. I've leaved it for an hour, nothing happen.

Reply
0 Kudos
NoeHoyos
Contributor
Contributor

Hello,

Ran into the same situation, very frustraiting. I upgraded to Workstation 10 (running on CentOS) and my Windows 2008 VMs crashed, however my old Workstation 6 Windows machines worked fine. After some time of troubleshoting I noticed it happened after installing vmware tools; VMs without the vmware tools installed worked fine. I decided to install VMware tools 8.6 on my Windows 2008 systems and have not seen a crash since then. I got the VMware tools from the link of this KB: http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=103681...

Thanks,

Noe

Reply
0 Kudos
sengork
Contributor
Contributor

Me too. RHEL 6.5 with Workstation 10.0.1 build-1379776.

Reply
0 Kudos
rileyb76
Enthusiast
Enthusiast

Any update on when Workstation 10.0.2 will be released?

Reply
0 Kudos
cscarff
Contributor
Contributor

I see VMware Workstation 10.0.2 is now ready for download, as of April 17th.

Chris
Reply
0 Kudos