After booting into the 2.6.22 kernel, recompiling vmware-modules and reconfiguring vmware-server, starting a new virtual machine will hang the system on AMD64. Several folks on the Gentoo Forums have reported the same issue. http://forums.gentoo.org/viewtopic-t-572276-highlight-vmware.html Reproducible: Always Steps to Reproduce: 1. Boot into 2.6.22 2. /etc/init.d/vmware start 3. vmware 4. Start a VM Actual Results: Everything hangs. Running processes continue to run but am unable to launch new processes.
Uhm... which vmware stuff doesn't run (exact ebuild name and version)? emerge --info output? Anything relevant in system logs? How are you VMs configured?
Created attachment 128270 [details] Log snippet while running crashed vmware
Created attachment 128271 [details] Output of emerge --info
Created attachment 128273 [details] VM configuration file The affected version is vmware-server, with the ebuild app-emulation/vmware-server-1.0.3.44356. I've tested with the attached configuration.
Reopen.
Hi, I'm very sorry, but this seems to be a well known problem (see http://lkml.org/lkml/2007/5/2/151). There is little we can do about this, as the solution proposed in the linked email was strongly advised against, and we won't be applying it. Currently you'll either have to use an old kernel, or wait until a new version of vmware-server is released. 5:(
*** Bug 189957 has been marked as a duplicate of this bug. ***
(In reply to comment #6) > Hi, I'm very sorry, but this seems to be a well known problem (see > http://lkml.org/lkml/2007/5/2/151). There is little we can do about this, as > the solution proposed in the linked email was strongly advised against, and we > won't be applying it. Currently you'll either have to use an old kernel, or > wait until a new version of vmware-server is released. 5:( Doesn't vmware-any-any-update113 fixes these issues?
Rodrigo, No, unfortunately any-any-update113 changelog says only "Change ASSERT_ON_COMPILE to make g++ 3.4 happy", which fixes the gcc-3 compilation issues, but not the AMD64 hard crash problem. Sorry... 5:(
*** Bug 192122 has been marked as a duplicate of this bug. ***
OK. Ramereth just affirmed this bug on VMware workstation, too. I'm going to REOPEN this until any-any114 comes out that should fix this so we know who has experienced the issue to have them help test the fix.
From what I could make out from the LKML it may not be possible for them to patch it properly for older versions of workstation without releasing new versions of the main software itself. I'm not aware that any-any-update114 will contain a fix, so this bug may end up being open a while...
Same here (2.6.23-rc* kernel). How about VMware Server 1.0.4? It may be resolved in this new version... http://www.vmware.com/support/server/doc/releasenotes_server.html#resolved
(In reply to comment #13) > Same here (2.6.23-rc* kernel). How about VMware Server 1.0.4? It may be > resolved in this new version... > http://www.vmware.com/support/server/doc/releasenotes_server.html#resolved > VMware Server 1.0.4 seems to have resolved this on 2.6.22-gentoo-r5.
(In reply to comment #14) > VMware Server 1.0.4 seems to have resolved this on 2.6.22-gentoo-r5. Same here (2.6.23-rc7). We can mark this bug as RESOLVED FIXED now, I guess.
I don't really see where in the resolved notes, or how it could have so please feel free to reopen this if necessary, but for now I'm gonna mark it as FIXED.
*** Bug 202925 has been marked as a duplicate of this bug. ***
I would like to request that this bug is reopened, as the problem is still present in vmware-workstation-5.5.5.56455 (the latest stable one for amd64). /Jakob
(In reply to comment #18) > I would like to request that this bug is reopened, as the problem is still > present in vmware-workstation-5.5.5.56455 (the latest stable one for amd64). Tried to re-emerge vmware-modules so that you get an uptodate any-any patch?
(In reply to comment #18) > I would like to request that this bug is reopened, as the problem is still > present in vmware-workstation-5.5.5.56455 (the latest stable one for amd64). Just to let you know I have several amd64 servers running app-emulation/vmware-server-1.0.4.56528 without problem.
I rebuilt vmware-modules 10 days ago. Should I use the unstable vmware-modules 1.0.0.17? Right now I am trying an evaluation license of version 6, but I would prefer not to have to pay for an upgrade. /Jakob
The thing that did the trick for me in this case was to remove any virtualization from the kernel. All kdm, xen and iguest support gone seems to make vmware happier. I havent tried this with workstation 5.5 though, just 6.0.
Jakob, please don't try vmware-modules-1.0.0.17, you'll get a version mismatch, 1.0.0.17 is only for vmware-workstation-6.0.2/vmware-player-2.0.2. The dependencies in the ebuild should ensure that you only use the correct vmware-modules for your package. Jakob, I'm happy to reopen this bug (although it will most likely just sit here, since there's very little we can do), but I'd like to request that you try out the latest 2.6.23 kernel with workstation-5.5.5 and see if that solves your problems? Thanks 5:)
(In reply to comment #23) > Jakob, please don't try vmware-modules-1.0.0.17, you'll get a version mismatch, > 1.0.0.17 is only for vmware-workstation-6.0.2/vmware-player-2.0.2. The > dependencies in the ebuild should ensure that you only use the correct > vmware-modules for your package. Happy New Year, and sorry for the late reply - I was away from the affected machine. I unmerged and remerged both vmware-workstation and vmware-modules. Unfortunately, the bug is still there both with the newest stable and the newest testing gentoo-sources kernels (2.6.23-r3 and -r5). I do not have any virtualization support in the kernel. When running Internet Explorer, the machines quickly crashes. This time it managed to pop up a window saying "vcpu-1:DoubleFault". When attempting to cut-and-paste the exact error into an editor, the whole machine crashed. > > Jakob, I'm happy to reopen this bug (although it will most likely just sit > here, since there's very little we can do), but I'd like to request that you > try out the latest 2.6.23 kernel with workstation-5.5.5 and see if that solves > your problems? > > Thanks 5:) > I am aware that reopening will not help much, but at least other people with the same problem are more likely to find it. To reiterate: I am running vmware-workstation-5.5.5.56455 and vmware-modules-1.0.0.15-r1 on an Intel Core 2 laptop with two processors and a 64-bit Linux installed. The guest operating system is Windows XP SP2 (32-bits). Best regards Jakob
Thanks for testing it out, re-opening it as requested.
(In reply to comment #25) > Thanks for testing it out, re-opening it as requested. > I upgraded to vmware-workstation 6, that seems to have solved the problem. /Jakob
(In reply to comment #26) > (In reply to comment #25) > > Thanks for testing it out, re-opening it as requested. > > > > I upgraded to vmware-workstation 6, that seems to have solved the problem. I'm perfectly happy with workstation 5, and have no need to upgrade. I resent being forced to pay an upgrade fee just to fix a bug, the introduction of which was none of my doing. (Sounds like something Microsoft would pull.) Is there any prospect of this being fixed any time soon? Please understand that I'm mad at vmware. The gentoo developers are doing a great job -- I have no complaint with them.
Stuart, I know it's along shot, but there have been some patches added to the vmware-modules to get them to run under the 2.6.25 kernel. Also vmware-workstation 5.5.6 has been released. Could you please try this combination (modules, workstation and preferably kernel) and report back how it works?
(In reply to comment #28) > Stuart, I know it's along shot, but there have been some patches added to the > vmware-modules to get them to run under the 2.6.25 kernel. Also > vmware-workstation 5.5.6 has been released. Could you please try this > combination (modules, workstation and preferably kernel) and report back how it > works? Well, I've already unmerged workstation from my x86_64 machine at work, but I still have it on my 32-bit machine at home, but I hardly use it. I was under the impression that it's a 64-bit problem. I can try it at home, if you think it will help. Also, the 2.6.25 kernel is still marked testing, isn't it?
You're right, only kernel 2.6.24 is available. It is an amd64 issue, so if it's too much trouble to test out then not to worry, I still don't know anything specific that's happened that might help, just wondering if vmware found a fix with their latest version. Sorry we haven't been able to find a solution...
(In reply to comment #30) > Sorry we haven't been able to find a > solution... Actually, my solution is to use VirtualBox. It works really well!
Ok, can someone please confirm whether the latest vmware-workstation-6.5.1 is still having the same issues on the latest kernel (2.6.28)?
