I built guest vm's one for Windows 7 and one for Windows XP using the virtual machine manager on a just updated to centos 5.7, and they are both crashing the host machine. They run only for a few minutes, but suddenly freeze, crashing the host. There is no networking. No X. No way to drop out of X. The only way out is a hard reboot. I don't see anything in the logs -- messages or libvirt logs -- immediately before the crash.
I haven't found anything like this on the web or on this list. The workstation has two xeon E5410s. I noticed that both the kvm-amd and kvm-intel modules are loaded, but don't know if that would cause a problem. I had an ati firepro graphics card in the machine, but suspected that might be the source of some conflict, and I put in an Nvidia card.
The vm's were built with all the defaults. The configuration is just about identical to vms I have running on a smaller machine with a dual core Athalon.
Thanks for any suggestions.
On 10/03/2011 10:16 PM, Negative wrote:
I built guest vm's one for Windows 7 and one for Windows XP using the virtual machine manager on a just updated to centos 5.7, and they are both crashing the host machine. They run only for a few minutes, but suddenly freeze, crashing the host. There is no networking. No X. No way to drop out of X. The only way out is a hard reboot. I don't see anything in the logs -- messages or libvirt logs -- immediately before the crash.
I haven't found anything like this on the web or on this list. The workstation has two xeon E5410s. I noticed that both the kvm-amd and kvm-intel modules are loaded, but don't know if that would cause a problem. I had an ati firepro graphics card in the machine, but suspected that might be the source of some conflict, and I put in an Nvidia card.
The vm's were built with all the defaults. The configuration is just about identical to vms I have running on a smaller machine with a dual core Athalon.
Thanks for any suggestions.
What hypervisor/version?
On Mon, Oct 3, 2011 at 10:16 PM, Negative negativebinomial@gmail.com wrote:
I built guest vm's one for Windows 7 and one for Windows XP using the virtual machine manager on a just updated to centos 5.7, and they are both crashing the host machine. They run only for a few minutes, but suddenly freeze, crashing the host. There is no networking. No X. No way to drop out of X. The only way out is a hard reboot. I don't see anything in the logs -- messages or libvirt logs -- immediately before the crash.
I haven't found anything like this on the web or on this list. The workstation has two xeon E5410s. I noticed that both the kvm-amd and kvm-intel modules are loaded, but don't know if that would cause a problem. I had an ati firepro graphics card in the machine, but suspected that might be the source of some conflict, and I put in an Nvidia card.
The vm's were built with all the defaults. The configuration is just about identical to vms I have running on a smaller machine with a dual core Athalon.
Thanks for any suggestions.
Is this new hardware? Have you run any hardware burn testing (CPU, RAM, etc...) and/or memtest86+ on the RAM? This sounds like a hardware issue to me.
-☙ Brian Mathis ❧-
Brian Mathis wrote:
On Mon, Oct 3, 2011 at 10:16 PM, Negative negativebinomial@gmail.com wrote:
I built guest vm's one for Windows 7 and one for Windows XP using the virtual machine manager on a  just updated to centos 5.7, and they are both crashing the host machine. They run only  for a few minutes, but suddenly freeze, crashing the host.  There is no networking. No X. No way to drop out of X. The only way out is a hard reboot. I don't see anything in the logs -- messages or libvirt logs -- immediately before the crash.
<snip>
Is this new hardware? Have you run any hardware burn testing (CPU, RAM, etc...) and/or memtest86+ on the RAM? This sounds like a hardware issue to me.
I agree with Brian - it may be coincidental that you built the VMs, and then it started crashing.
One other question: is selinux enabled?
mark
On Tue, Oct 4, 2011 at 10:41 AM, m.roth@5-cent.us wrote:
Brian Mathis wrote:
On Mon, Oct 3, 2011 at 10:16 PM, Negative negativebinomial@gmail.com wrote:
I built guest vm's one for Windows 7 and one for Windows XP using the virtual machine manager on a  just updated to centos 5.7, and they are both crashing the host machine. They run only  for a few minutes, but suddenly freeze, crashing the host.  There is no networking. No X. No way to drop out of X. The only way out is a hard reboot. I don't see anything in the logs -- messages or libvirt logs -- immediately before the crash.
<snip> > Is this new hardware? Have you run any hardware burn testing (CPU, > RAM, etc...) and/or memtest86+ on the RAM? This sounds like a > hardware issue to me.
It's about three years old. I had one hardware issue a year ago in which a video card fried, but it's been great. I will run memtest this afternoon.
I agree with Brian - it may be coincidental that you built the VMs, and then it started crashing.
I should run memtest. I don't know of a tool to check the processors. I use the machine for analyzing data, and often use most of the 32 gigs of memory in it, but I doubt I've ever seriously stressed the processors.
I created the two guests with the gui, but since they crash, I started one without starting X on the host, using virsh. The guest and host both stay up. When starting using virsh with the --console switch I get what looks like a telnet connection. But I know almost nothing about Windows and don't know what to look at. Networking between the guest and host might be borked -- and that would've been my fault. Then, every time X is running the guest and host crash.
One other question: is selinux enabled?
Yes. No warnings, though.
mark
On Tue, Oct 4, 2011 at 12:05 PM, Negative negativebinomial@gmail.com wrote:
On Tue, Oct 4, 2011 at 10:41 AM, m.roth@5-cent.us wrote:
Brian Mathis wrote:
On Mon, Oct 3, 2011 at 10:16 PM, Negative negativebinomial@gmail.com wrote:
I built guest vm's one for Windows 7 and one for Windows XP using the virtual machine manager on a  just updated to centos 5.7, and they are both crashing the host machine. They run only  for a few minutes, but suddenly freeze, crashing the host.  There is no networking. No X. No way to drop out of X. The only way out is a hard reboot. I don't see anything in the logs -- messages or libvirt logs -- immediately before the crash.
<snip> > Is this new hardware? Have you run any hardware burn testing (CPU, > RAM, etc...) and/or memtest86+ on the RAM? This sounds like a > hardware issue to me.
It's about three years old. I had one hardware issue a year ago in which a video card fried, but it's been great. I will run memtest this afternoon.
I agree with Brian - it may be coincidental that you built the VMs, and then it started crashing.
I should run memtest. I don't know of a tool to check the processors. I use the machine for analyzing data, and often use most of the 32 gigs of memory in it, but I doubt I've ever seriously stressed the processors.
I created the two guests with the gui, but since they crash, I started one without starting X on the host, using virsh. The guest and host both stay up. When starting using virsh with the --console switch I get what looks like a telnet connection. But I know almost nothing about Windows and don't know what to look at. Networking between the guest and host might be borked -- and that would've been my fault. Then, every time X is running the guest and host crash.
One other question: is selinux enabled?
Yes. No warnings, though.
mark
It should not matter what the guest is, so Windows or Linux it shouldn't be crashing. If not hardware, it points to a bug in the hypervisor software.
-☙ Brian Mathis ❧-
Negative wrote:
On Tue, Oct 4, 2011 at 10:41 AM, m.roth@5-cent.us wrote:
Brian Mathis wrote:
On Mon, Oct 3, 2011 at 10:16 PM, Negative negativebinomial@gmail.com wrote:
I built guest vm's one for Windows 7 and one for Windows XP using the virtual machine manager on a  just updated to centos 5.7, and they are both crashing the host machine. They run only  for a few minutes,
<snip>
Is this new hardware? Have you run any hardware burn testing (CPU, RAM, etc...) and/or memtest86+ on the RAM? This sounds like a hardware issue to me.
It's about three years old. I had one hardware issue a year ago in which a video card fried, but it's been great. I will run memtest this afternoon.
I agree with Brian - it may be coincidental that you built the VMs, and then it started crashing.
<snip>
-- and that would've been my fault. Then, every time X is running the guest and host crash.
One other question: is selinux enabled?
Yes. No warnings, though.
More and more it sounds like a hardware issue. Hmm, every time X is running, and you say you had one video card fried - how did it fry? Also, is this machine on a good quality surge protector? Have you had a thunderstorm, or power outages recently?
mark
On Tue, Oct 4, 2011 at 1:18 PM, m.roth@5-cent.us wrote:
Negative wrote:
On Tue, Oct 4, 2011 at 10:41 AM, m.roth@5-cent.us wrote:
Brian Mathis wrote:
On Mon, Oct 3, 2011 at 10:16 PM, Negative <negativebinomial@gmail.com
wrote:
I built guest vm's one for Windows 7 and one for Windows XP using the virtual machine manager on a  just updated to centos 5.7, and they are both crashing the host machine. They run only  for a few minutes,
<snip> >> > Is this new hardware? Have you run any hardware burn testing (CPU, >> > RAM, etc...) and/or memtest86+ on the RAM? This sounds like a >> > hardware issue to me. > > It's about three years old. I had one hardware issue a year ago in which a > video card fried, but it's been great. I will run memtest this afternoon. > >> I agree with Brian - it may be coincidental that you built the VMs, and >> then it started crashing. <snip> > -- and that would've been my fault. Then, every time X is running the > guest and host crash. >> >> One other question: is selinux enabled? > > Yes. No warnings, though.
More and more it sounds like a hardware issue. Hmm, every time X is running, and you say you had one video card fried - how did it fry? Also, is this machine on a good quality surge protector? Have you had a thunderstorm, or power outages recently?
mark
The vendor told me that the particular video card model (I forget which) had some flaw. In any case the fan stopped running, and it heated up. I usually use the machine remotely but I was at the console at that moment. The monitor started flickering and then went gray.
The vendor sent a replacement, but I had thrown an old ATI in before it arrived. When the crashes occurred now, I finally put in the replacement. Same behavior.
The surge protector is good. I live in NYC and the biggest environmental hazard is the cleaning lady, who has in the past tripped the surge protector switch.
I fear you're right about the hardware. But as far as I can tell everything else works fine. I went overboard in buying two quad processors -- so I could live with one if that's the problem.
Negative wrote:
On Tue, Oct 4, 2011 at 1:18 PM, m.roth@5-cent.us wrote:
Negative wrote:
On Tue, Oct 4, 2011 at 10:41 AM, m.roth@5-cent.us wrote:
Brian Mathis wrote:
On Mon, Oct 3, 2011 at 10:16 PM, Negative
<negativebinomial@gmail.com
wrote:
I built guest vm's one for Windows 7 and one for Windows XP using the virtual machine manager on a  just updated to centos 5.7, and they are both crashing the host machine. They run only  for a few
minutes,
<snip> >> > Is this new hardware? Have you run any hardware burn testing (CPU, >> > RAM, etc...) and/or memtest86+ on the RAM? This sounds like a >> > hardware issue to me. >
<snip>
More and more it sounds like a hardware issue. Hmm, every time X is running, and you say you had one video card fried - how did it fry? Also, is this machine on a good quality surge protector? Have you had a thunderstorm, or power outages recently?
The vendor told me that the particular video card model (I forget which) had some flaw. In any case the fan stopped running, and it heated up. I usually use the machine remotely but I was at the console at that moment. The monitor started flickering and then went gray.
The vendor sent a replacement, but I had thrown an old ATI in before it arrived. When the crashes occurred now, I finally put in the replacement. Same behavior.
Have you examined the m/b and cards *around* where the card fried? Its heat death may have affected things around it.
The surge protector is good. I live in NYC and the biggest environmental hazard is the cleaning lady, who has in the past tripped the surge protector switch.
<g> Do you know the story about the mainframe shop, the racks of tapes, and the cleaning staff?
I fear you're right about the hardware. But as far as I can tell everything else works fine. I went overboard in buying two quad processors -- so I could live with one if that's the problem.
A replacement m/b?
mark
On Tue, Oct 4, 2011 at 2:24 PM, m.roth@5-cent.us wrote:
Negative wrote:
On Tue, Oct 4, 2011 at 1:18 PM, m.roth@5-cent.us wrote:
Negative wrote:
On Tue, Oct 4, 2011 at 10:41 AM, m.roth@5-cent.us wrote:
Brian Mathis wrote:
On Mon, Oct 3, 2011 at 10:16 PM, Negative
<negativebinomial@gmail.com
wrote: > I built guest vm's one for Windows 7 and one for Windows XP using > the > virtual machine manager on a  just updated to centos 5.7, and they > are both crashing the host machine. They run only  for a few
minutes,
<snip> >> > Is this new hardware? Have you run any hardware burn testing (CPU, >> > RAM, etc...) and/or memtest86+ on the RAM? This sounds like a >> > hardware issue to me. >
<snip> >> More and more it sounds like a hardware issue. Hmm, every time X is >> running, and you say you had one video card fried - how did it fry? >> Also, >> is this machine on a good quality surge protector? Have you had a >> thunderstorm, or power outages recently? > > The vendor told me that the particular video card model (I forget which) > had some flaw. In any case the fan stopped running, and it heated up. I > usually use the machine remotely but I was at the console at that moment. > The monitor started flickering and then went gray. > > The vendor sent a replacement, but I had thrown an old ATI in before it > arrived. When the crashes occurred now, I finally put in the replacement. > Same behavior.
Have you examined the m/b and cards *around* where the card fried? Its heat death may have affected things around it.
It looks good to the eye. The capacitors look good.
The surge protector is good. I live in NYC and the biggest environmental hazard is the cleaning lady, who has in the past tripped the surge protector switch.
<g> Do you know the story about the mainframe shop, the racks of tapes, and the cleaning staff?
Don't know it but I can imagine.
I fear you're right about the hardware. But as far as I can tell everything else works fine. I went overboard in buying two quad processors -- so I could live with one if that's the problem.
A replacement m/b?
That's a tough one!
Since the crashes can be duplicated and are only caused by this one combination of events, I don't know.
On another machine, I had a case where none of the kvm guests would boot. It turned out to be a conflict between libvirt and the nvidia proprietary driver. I used an old version of the video driver until Nvidia caught up. (It only affected amd processors.)
On Tue, Oct 4, 2011 at 2:51 PM, Negative negativebinomial@gmail.com wrote:
Since the crashes can be duplicated and are only caused by this one combination of events, I don't know.
On another machine, I had a case where none of the kvm guests would boot. It turned out to be a conflict between libvirt and the nvidia proprietary driver. I used an old version of the video driver until Nvidia caught up. (It only affected amd processors.)
Do you have to use the local video at all? What happens if you use freenx and run in a session started from a remote NX client?
On Tue, Oct 4, 2011 at 4:16 PM, Les Mikesell lesmikesell@gmail.com wrote:
On Tue, Oct 4, 2011 at 2:51 PM, Negative negativebinomial@gmail.com wrote:
Since the crashes can be duplicated and are only caused by this one combination of events, I don't know.
On another machine, I had a case where none of the kvm guests would boot.
It
turned out to be a conflict between libvirt and the nvidia proprietary driver. I used an old version of the video driver until Nvidia caught up. (It only affected amd processors.)
Do you have to use the local video at all? What happens if you use freenx and run in a session started from a remote NX client?
Doesn't freenx use X? I haven't installed but will try it.
What I did try was to leave X off on workstation and built a fedora guest. I also connected to the workstation from another machine via ssh and I tried running the virt-viewer via X-forwarding, it stayed up for about 10 or 15 minutes and crashed. Since I could see the console on the work station then, it showed a kernel panic.
After rebooting I opened a terminal via ssh and it seems ok -- up for about 30 minutes so far, I'll check in the morning.
The trouble with that is I usually connect to the workshop via vnc, so I'm not sure the guest will be of much use.
On Tue, Oct 4, 2011 at 11:41 PM, Negative negativebinomial@gmail.com wrote:
Do you have to use the local video at all? What happens if you use freenx and run in a session started from a remote NX client?
Doesn't freenx use X? I haven't installed but will try it.
Yes, it uses X, but it doesn't use the local video hardware.
What I did try was to leave X off on workstation and built a fedora guest. I also connected to the workstation from another machine via ssh and I tried running the virt-viewer via X-forwarding, it stayed up for about 10 or 15 minutes and crashed. Since I could see the console on the work station then, it showed a kernel panic.
Not sure what that means. Freenx would at least keep the sesson active when you disconnect.
After rebooting I opened a terminal via ssh and it seems ok -- up for about 30 minutes so far, I'll check in the morning.
The trouble with that is I usually connect to the workshop via vnc, so I'm not sure the guest will be of much use.
A vncserver session not attached to the local console should also avoid local hardware issues - but freenx/NX is nicer to use.
On Wed, Oct 5, 2011 at 1:10 AM, Les Mikesell lesmikesell@gmail.com wrote:
On Tue, Oct 4, 2011 at 11:41 PM, Negative negativebinomial@gmail.com wrote:
Do you have to use the local video at all? What happens if you use freenx and run in a session started from a remote NX client?
Doesn't freenx use X? I haven't installed but will try it.
Yes, it uses X, but it doesn't use the local video hardware.
What I did try was to leave X off on workstation and built a fedora
guest. I
also connected to the workstation from another machine via ssh and I
tried
running the virt-viewer via X-forwarding, it stayed up for about 10 or 15 minutes and crashed. Since I could see the console on the work station
then,
it showed a kernel panic.
Not sure what that means. Freenx would at least keep the sesson active when you disconnect.
I was trying to say, but not very well, that when qemu-kvm is running a guest and X is running, whether it's on the local video hardware or not, I get a kernel panic.
So now I've found I can run X without problems and I can run qemu-kvm without problems, but I cannot run both at the same time.
I wonder if an older version of the kvm package would work around this, or maybe an older version of the kernel.
kernel: 2..18-274.3.1.el5 kvm: 83-239.el5.centos
After rebooting I opened a terminal via ssh and it seems ok -- up for
about
30 minutes so far, I'll check in the morning.
The trouble with that is I usually connect to the workshop via vnc, so I'm not sure the guest will be of much use.
A vncserver session not attached to the local console should also
avoid local hardware issues - but freenx/NX is nicer to use.
-- Les Mikesell lesmikesell@gmail.com _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
On Wed, Oct 5, 2011 at 8:17 AM, Negative negativebinomial@gmail.com wrote:
Not sure what that means. Freenx would at least keep the sesson active when you disconnect.
I was trying to say, but not very well, that when qemu-kvm is running a guest and X is running, whether it's on the local video hardware or not, I get a kernel panic.
So now I've found I can run X without problems and I can run qemu-kvm without problems, but I cannot run both at the same time.
'X' should be split cleanly into client and server programs where in X-speak the server serves the display and keyboard and programs are the clients. An X client program should not be touching any hardware directly. But, virtualization stuff might try to cheat. If you don't need the local console, you might try loading VMware ESXi first, then run all your other OS's as guests under that - you do need a windows box to run as the console when making changes or installing things, though.
Les Mikesell wrote:
On Wed, Oct 5, 2011 at 8:17 AM, Negative negativebinomial@gmail.com wrote:
<snip>
So now I've found I can run X without problems and I can run qemu-kvm without problems, but I cannot run both at the same time.
'X' should be split cleanly into client and server programs where in X-speak the server serves the display and keyboard and programs are the clients. An X client program should not be touching any hardware directly. But, virtualization stuff might try to cheat. If you don't need the local console, you might try loading VMware ESXi first, then run all your other OS's as guests under that - you do need a windows box to run as the console when making changes or installing things, though.
Actually, if you do that, you can log into ESX1 - it's actually a modified RHEL 3, I think. It's the remote admin GUI that you need WinDoze for.
mark
On Wed, Oct 5, 2011 at 11:00 AM, m.roth@5-cent.us wrote:
Les Mikesell wrote:
On Wed, Oct 5, 2011 at 8:17 AM, Negative negativebinomial@gmail.com wrote:
<snip> >> So now I've found I can run X without problems and I can run qemu-kvm >> without problems, but I cannot run both at the same time. > > 'X' should be split cleanly into client and server programs where in > X-speak the server serves the display and keyboard and programs are > the clients. An X client program should not be touching any hardware > directly. But, virtualization stuff might try to cheat. If you > don't need the local console, you might try loading VMware ESXi first, > then run all your other OS's as guests under that - you do need a > windows box to run as the console when making changes or installing > things, though.
Actually, if you do that, you can log into ESX1 - it's actually a modified RHEL 3, I think. It's the remote admin GUI that you need WinDoze for.
mark
VMware's a possibility. Would I need to remove the kvm package? Do I have to keep libvirtd off?
I still wonder what is causing this. I couldn't find any mention of a similar problem, including on my desktop in my office, where I have a very similar setup, with four kvm guests, two Fedora, one Centos 6 and one Windows XP.
Negative wrote:
On Wed, Oct 5, 2011 at 11:00 AM, m.roth@5-cent.us wrote:
Les Mikesell wrote:
On Wed, Oct 5, 2011 at 8:17 AM, Negative negativebinomial@gmail.com wrote:
<snip> >> So now I've found I can run X without problems and I can run qemu-kvm >> without problems, but I cannot run both at the same time. > > 'X' should be split cleanly into client and server programs where in > X-speak the server serves the display and keyboard and programs are > the clients. An X client program should not be touching any hardware > directly. But, virtualization stuff might try to cheat. If you > don't need the local console, you might try loading VMware ESXi first, > then run all your other OS's as guests under that - you do need a > windows box to run as the console when making changes or installing > things, though.
Actually, if you do that, you can log into ESX1 - it's actually a modified RHEL 3, I think. It's the remote admin GUI that you need WinDoze for.
VMware's a possibility. Would I need to remove the kvm package? Do I have to keep libvirtd off?
You could.
I still wonder what is causing this. I couldn't find any mention of a similar problem, including on my desktop in my office, where I have a very similar setup, with four kvm guests, two Fedora, one Centos 6 and one Windows XP.
It keeps coming back to sounding like a hardware problem, maybe the video card.
mark
On Wed, Oct 5, 2011 at 11:15 AM, m.roth@5-cent.us wrote:
Negative wrote:
On Wed, Oct 5, 2011 at 11:00 AM, m.roth@5-cent.us wrote:
Les Mikesell wrote:
On Wed, Oct 5, 2011 at 8:17 AM, Negative negativebinomial@gmail.com wrote:
<snip> >> So now I've found I can run X without problems and I can run qemu-kvm >> without problems, but I cannot run both at the same time. > > 'X' should be split cleanly into client and server programs where in > X-speak the server serves the display and keyboard and programs are > the clients. An X client program should not be touching any hardware > directly. But, virtualization stuff might try to cheat. If you > don't need the local console, you might try loading VMware ESXi first, > then run all your other OS's as guests under that - you do need a > windows box to run as the console when making changes or installing > things, though.
Actually, if you do that, you can log into ESX1 - it's actually a modified RHEL 3, I think. It's the remote admin GUI that you need WinDoze for.
VMware's a possibility. Would I need to remove the kvm package? Do I
have
to keep libvirtd off?
You could.
I still wonder what is causing this. I couldn't find any mention of a similar problem, including on my desktop in my office, where I have a
very
similar setup, with four kvm guests, two Fedora, one Centos 6 and one Windows XP.
It keeps coming back to sounding like a hardware problem, maybe the video card.
mark
That was my first thought. I've had the same behavior with two video cards -- an ATI and an nvidia.
Could it be that the kvm-amd module causes problems? It is loaded along with the kvm-intel and kvm. .
Negative wrote:
On Wed, Oct 5, 2011 at 11:15 AM, m.roth@5-cent.us wrote:
Negative wrote:
On Wed, Oct 5, 2011 at 11:00 AM, m.roth@5-cent.us wrote:
Les Mikesell wrote:
On Wed, Oct 5, 2011 at 8:17 AM, Negative
wrote:
<snip> >> So now I've found I can run X without problems and I can run >> qemu-kvm without problems, but I cannot run both at the same time.
<snip>
I still wonder what is causing this. I couldn't find any mention of a
<snip>
It keeps coming back to sounding like a hardware problem, maybe the video card.
That was my first thought. I've had the same behavior with two video cards -- an ATI and an nvidia.
Could it be that the kvm-amd module causes problems? It is loaded along with the kvm-intel and kvm. .
If you've had the *same* behaviour with two separate video cards, from two different vendors, then I start wondering about either the m/b, or memory.
mark
Negative wrote:
On Wed, Oct 5, 2011 at 11:15 AM, m.roth@5-cent.us wrote: Negative wrote: On Wed, Oct 5, 2011 at 11:15 AM, m.roth@5-cent.us wrote:
Negative wrote:
<snip>
I still wonder what is causing this. I couldn't find any mention of a similar problem, including on my desktop in my office, where I have a very similar setup, with four kvm guests, two Fedora, one Centos 6 and one Windows XP.
<snip> Do I remember this is 5.7? Look at the announcement that *just* came out in the last hour, with the libX11 bugfix. https://rhn.redhat.com/errata/RHBA-2011-1351.html says "Previously, in the 64-bit mode, libX11 computed addresses using the 32-bit arithmetic. As a consequence, under heavy load, applications running in the X environment terminated unexpectedly. A patch has been provided to address this issue, and the crashes no longer occur in the described scenario."
mark
On Wed, Oct 5, 2011 at 1:00 PM, m.roth@5-cent.us wrote:
Negative wrote:
On Wed, Oct 5, 2011 at 11:15 AM, m.roth@5-cent.us wrote: Negative wrote: On Wed, Oct 5, 2011 at 11:15 AM, m.roth@5-cent.us wrote:
Negative wrote:
<snip> >> > I still wonder what is causing this. I couldn't find any mention of a >> > similar problem, including on my desktop in my office, where I have a >> > very similar setup, with four kvm guests, two Fedora, one Centos 6 and >> > one Windows XP. <snip> Do I remember this is 5.7? Look at the announcement that *just* came out in the last hour, with the libX11 bugfix. <https://rhn.redhat.com/errata/RHBA-2011-1351.html> says "Previously, in the 64-bit mode, libX11 computed addresses using the 32-bit arithmetic. As a consequence, under heavy load, applications running in the X environment terminated unexpectedly. A patch has been provided to address this issue, and the crashes no longer occur in the described scenario."
mark
If this isn't my lucky day. RH and Centos solved my problem even before I defined it.
I saw the update earlier and didn't dare hope. I updated and it seems to have solved the issue. On the host machine, I fired up virt-manager, started the Fedora guest and it's been up for a half hour.
Now I, too, can start complaining about Gnome 3. I've read it's like Windows, but it's the spitting image of the Mac OS.
Negative wrote: <MVNCH>
Now I, too, can start complaining about Gnome 3. I've read it's like Windows, but it's the spitting image of the Mac OS.
Maybe, but I was having to deal with it on a user's fedora 15 machine, and I ACTIVELY dislike it, with it's scroll-over-and-the-transparent-menu-fades-in crap.
Pointless and annoying eye candy.
mark, getting back to real work on his command lines
On Wed, Oct 5, 2011 at 3:06 PM, Negative negativebinomial@gmail.com wrote:
On Wed, Oct 5, 2011 at 1:00 PM, m.roth@5-cent.us wrote:
Negative wrote:
On Wed, Oct 5, 2011 at 11:15 AM, m.roth@5-cent.us wrote: Negative wrote: On Wed, Oct 5, 2011 at 11:15 AM, m.roth@5-cent.us wrote:
Negative wrote:
<snip> >> > I still wonder what is causing this. I couldn't find any mention of a >> > similar problem, including on my desktop in my office, where I have a >> > very similar setup, with four kvm guests, two Fedora, one Centos 6 and >> > one Windows XP. <snip> Do I remember this is 5.7? Look at the announcement that *just* came out in the last hour, with the libX11 bugfix. <https://rhn.redhat.com/errata/RHBA-2011-1351.html> says "Previously, in the 64-bit mode, libX11 computed addresses using the 32-bit arithmetic. As a consequence, under heavy load, applications running in the X environment terminated unexpectedly. A patch has been provided to address this issue, and the crashes no longer occur in the described scenario."
mark
And, Mark, thanks for mentioning it.
If this isn't my lucky day. RH and Centos solved my problem even before I defined it.
I saw the update earlier and didn't dare hope. I updated and it seems to have solved the issue. On the host machine, I fired up virt-manager, started the Fedora guest and it's been up for a half hour.
Now I, too, can start complaining about Gnome 3. I've read it's like Windows, but it's the spitting image of the Mac OS.
On Wed, Oct 5, 2011 at 3:11 PM, Negative negativebinomial@gmail.com wrote:
On Wed, Oct 5, 2011 at 3:06 PM, Negative negativebinomial@gmail.comwrote:
On Wed, Oct 5, 2011 at 1:00 PM, m.roth@5-cent.us wrote:
Negative wrote:
On Wed, Oct 5, 2011 at 11:15 AM, m.roth@5-cent.us wrote: Negative wrote: On Wed, Oct 5, 2011 at 11:15 AM, m.roth@5-cent.us wrote:
Negative wrote:
<snip> >> > I still wonder what is causing this. I couldn't find any mention of a >> > similar problem, including on my desktop in my office, where I have a >> > very similar setup, with four kvm guests, two Fedora, one Centos 6 and >> > one Windows XP. <snip> Do I remember this is 5.7? Look at the announcement that *just* came out in the last hour, with the libX11 bugfix. <https://rhn.redhat.com/errata/RHBA-2011-1351.html> says "Previously, in the 64-bit mode, libX11 computed addresses using the 32-bit arithmetic. As a consequence, under heavy load, applications running in the X environment terminated unexpectedly. A patch has been provided to address this issue, and the crashes no longer occur in the described scenario."
markAnd, Mark, thanks for mentioning it.
If this isn't my lucky day. RH and Centos solved my problem even before I defined it.
I saw the update earlier and didn't dare hope. I updated and it seems to have solved the issue. On the host machine, I fired up virt-manager, started the Fedora guest and it's been up for a half hour.
Now I, too, can start complaining about Gnome 3. I've read it's like Windows, but it's the spitting image of the Mac OS.
I spoke too soon. Crashed again after being up for several hours. I'm running memtest86 now.
Negative wrote:
On Wed, Oct 5, 2011 at 11:00 AM, m.roth@5-cent.us wrote:
Les Mikesell wrote:
On Wed, Oct 5, 2011 at 8:17 AM, Negative negativebinomial@gmail.com wrote:
<snip> >> So now I've found I can run X without problems and I can run qemu-kvm >> without problems, but I cannot run both at the same time. > > 'X' should be split cleanly into client and server programs where in > X-speak the server serves the display and keyboard and programs are > the clients. An X client program should not be touching any hardware > directly. But, virtualization stuff might try to cheat. If you > don't need the local console, you might try loading VMware ESXi first, > then run all your other OS's as guests under that - you do need a > windows box to run as the console when making changes or installing > things, though.
Actually, if you do that, you can log into ESX1 - it's actually a modified RHEL 3, I think. It's the remote admin GUI that you need WinDoze for.
VMware's a possibility. Would I need to remove the kvm package? Do I have to keep libvirtd off?
You could.
I still wonder what is causing this. I couldn't find any mention of a similar problem, including on my desktop in my office, where I have a very similar setup, with four kvm guests, two Fedora, one Centos 6 and one Windows XP.
It keeps coming back to sounding like a hardware problem, maybe the video card.
mark
Negative wrote:
On Wed, Oct 5, 2011 at 11:00 AM, m.roth@5-cent.us wrote:
Les Mikesell wrote:
On Wed, Oct 5, 2011 at 8:17 AM, Negative negativebinomial@gmail.com wrote:
<snip> >> So now I've found I can run X without problems and I can run qemu-kvm >> without problems, but I cannot run both at the same time. > > 'X' should be split cleanly into client and server programs where in > X-speak the server serves the display and keyboard and programs are > the clients. An X client program should not be touching any hardware > directly. But, virtualization stuff might try to cheat. If you > don't need the local console, you might try loading VMware ESXi first, > then run all your other OS's as guests under that - you do need a > windows box to run as the console when making changes or installing > things, though.
Actually, if you do that, you can log into ESX1 - it's actually a modified RHEL 3, I think. It's the remote admin GUI that you need WinDoze for.
VMware's a possibility. Would I need to remove the kvm package? Do I have to keep libvirtd off?
You could.
I still wonder what is causing this. I couldn't find any mention of a similar problem, including on my desktop in my office, where I have a very similar setup, with four kvm guests, two Fedora, one Centos 6 and one Windows XP.
It keeps coming back to sounding like a hardware problem, maybe the video card.
mark
On Wed, Oct 5, 2011 at 10:11 AM, Negative negativebinomial@gmail.com wrote:
VMware's a possibility. Would I need to remove the kvm package? Do I have to keep libvirtd off?
You could use VMware server or player as an alternative to kvm under Centos. VMware ESXi loads as the base OS, so you have to wipe everything and start over. The latest release (5.0) claims that virtualization can be 'nested' - that is you could run kvm (etc.) on a guest, although performance probably wouldn't be great.
Les Mikesell wrote:
On Wed, Oct 5, 2011 at 10:11 AM, Negative negativebinomial@gmail.com wrote:
VMware's a possibility. Would I need to remove the kvm package? Do I have to keep libvirtd off?
You could use VMware server or player as an alternative to kvm under Centos. VMware ESXi loads as the base OS, so you have to wipe everything and start over. The latest release (5.0) claims that virtualization can be 'nested' - that is you could run kvm (etc.) on a guest, although performance probably wouldn't be great.
Any idea what the new version is based on? Is it still a 2.4 kernel (and based on RHEL 3)?
mark
On Wed, Oct 5, 2011 at 11:02 AM, m.roth@5-cent.us wrote:
You could use VMware server or player as an alternative to kvm under Centos. VMware ESXi loads as the base OS, so you have to wipe everything and start over. The latest release (5.0) claims that virtualization can be 'nested' - that is you could run kvm (etc.) on a guest, although performance probably wouldn't be great.
Any idea what the new version is based on? Is it still a 2.4 kernel (and based on RHEL 3)?
Don't think they admit to any relationship to Linux. uname just says VMkernel 5.0.0 #1 with a build number. Most of the user level programs are implemented with busybox. It has ssh/scp, but no rsync.
On Wed, Oct 05, 2011 at 11:31:29AM -0500, Les Mikesell wrote:
On Wed, Oct 5, 2011 at 11:02 AM, m.roth@5-cent.us wrote:
Centos. ?VMware ESXi loads as the base OS, so you have to wipe everything and start over. ?The latest release (5.0) claims that virtualization can be 'nested' - that is you could run kvm (etc.) on a guest, although performance probably wouldn't be great.
Any idea what the new version is based on? Is it still a 2.4 kernel (and based on RHEL 3)?
Don't think they admit to any relationship to Linux. uname just says VMkernel 5.0.0 #1 with a build number. Most of the user level programs are implemented with busybox. It has ssh/scp, but no rsync.
Note the "i" in ESXi; this is an "embedded linux" variant. ESX is still RedHat based, to the best of my knowledge.
Les Mikesell wrote:
On Wed, Oct 5, 2011 at 11:02 AM, m.roth@5-cent.us wrote:
You could use VMware server or player as an alternative to kvm under Centos. VMware ESXi loads as the base OS, so you have to wipe everything and start over. The latest release (5.0) claims that virtualization can be 'nested' - that is you could run kvm (etc.) on a guest, although performance probably wouldn't be great.
Any idea what the new version is based on? Is it still a 2.4 kernel (and based on RHEL 3)?
Don't think they admit to any relationship to Linux. uname just says VMkernel 5.0.0 #1 with a build number. Most of the user level programs are implemented with busybox. It has ssh/scp, but no rsync.
It was - the docs for the version that was around in early '09 *said* that the latest release was based on RHEL 3.
mark
On Wed, Oct 5, 2011 at 12:01 PM, m.roth@5-cent.us wrote:
It was - the docs for the version that was around in early '09 *said* that the latest release was based on RHEL 3.
You are supposed to be able to download the open source components here: http://downloads.vmware.com/d/info/datacenter_cloud_infrastructure/vmware_vs... but all I see is a license file - that does mention the 2.6 kernel and drivers.
Les Mikesell wrote:
On Wed, Oct 5, 2011 at 12:01 PM, m.roth@5-cent.us wrote:
It was - the docs for the version that was around in early '09 *said* that the latest release was based on RHEL 3.
You are supposed to be able to download the open source components here: http://downloads.vmware.com/d/info/datacenter_cloud_infrastructure/vmware_vs... but all I see is a license file - that does mention the 2.6 kernel and drivers.
Yeah, well, the RHEL stuff I assume is released; their own heavy enhancements, I don't know.
mark
On Wed, Oct 5, 2011 at 12:16 PM, m.roth@5-cent.us wrote:
It was - the docs for the version that was around in early '09 *said* that the latest release was based on RHEL 3.
You are supposed to be able to download the open source components here: http://downloads.vmware.com/d/info/datacenter_cloud_infrastructure/vmware_vs... but all I see is a license file - that does mention the 2.6 kernel and drivers.
Yeah, well, the RHEL stuff I assume is released; their own heavy enhancements, I don't know.
It doesn't matter if someone else released source. Anyone distributing binaries containing GPL code is supposed to also supply the matching source along with anything else that becomes part of a derived work.
Les Mikesell wrote:
On Wed, Oct 5, 2011 at 12:16 PM, m.roth@5-cent.us wrote:
It was - the docs for the version that was around in early '09 *said* that the latest release was based on RHEL 3.
You are supposed to be able to download the open source components here: http://downloads.vmware.com/d/info/datacenter_cloud_infrastructure/vmware_vs... but all I see is a license file - that does mention the 2.6 kernel and drivers.
Yeah, well, the RHEL stuff I assume is released; their own heavy enhancements, I don't know.
It doesn't matter if someone else released source. Anyone distributing binaries containing GPL code is supposed to also supply the matching source along with anything else that becomes part of a derived work.
IIRC, I don't *think* that if you take it and enhance it, you're required to release your commercial enhancements. For example, video drivers, proprietary.
mark
On Wed, Oct 5, 2011 at 12:36 PM, m.roth@5-cent.us wrote:
Yeah, well, the RHEL stuff I assume is released; their own heavy enhancements, I don't know.
It doesn't matter if someone else released source. Anyone distributing binaries containing GPL code is supposed to also supply the matching source along with anything else that becomes part of a derived work.
IIRC, I don't *think* that if you take it and enhance it, you're required to release your commercial enhancements. For example, video drivers, proprietary.
The whole point of the GPL is to require the release of source (to anyone who gets binaries) of any derived work. Kernel modules aren't strictly considered to be derived from the kernel, although there has been some speculation that they could be.
On 2011-10-05, at 12:14 PM, Les Mikesell lesmikesell@gmail.com wrote:
On Wed, Oct 5, 2011 at 12:36 PM, m.roth@5-cent.us wrote:
Yeah, well, the RHEL stuff I assume is released; their own heavy enhancements, I don't know.
It doesn't matter if someone else released source. Anyone distributing binaries containing GPL code is supposed to also supply the matching source along with anything else that becomes part of a derived work.
IIRC, I don't *think* that if you take it and enhance it, you're required to release your commercial enhancements. For example, video drivers, proprietary.
The whole point of the GPL is to require the release of source (to anyone who gets binaries) of any derived work. Kernel modules aren't strictly considered to be derived from the kernel, although there has been some speculation that they could be.
VMware ESXi is available for free download -I think it's restricted to non-commercial use - but you have to register with VMware first. The VSphere Vcenter console is too.
I have it running on an Intel hardware based home server with half a dozen vm's. It does look and feel a lot like RHEL.
Gordon
On 10/05/11 11:14 AM, Les Mikesell wrote:
The whole point of the GPL is to require the release of source (to anyone who gets binaries) of any derived work. Kernel modules aren't strictly considered to be derived from the kernel, although there has been some speculation that they could be.
the closed source drivers usually have two pieces, clearly seperated. one piece is open source and interfaces with the kernel, the other piece implements the hardware specific features (be they wifi or 3D graphics or what) and is called from the 1st piece, and is supplied as a closed binary module.
its my understanding that there's NO traces of any linux kernel in ESXi. the RHEL that was in the original ESX was the management console, which ran in VM0 (much like dom0 on a Xen system), and wasn't the hypervisor. What I saw poking around a esxi4 install looked more akin to a stripped down BSD, with a BusyBox shell.
If you haven't already, check the mainboard & power supply for bad capacitors:
-
Brian Mathis -
Digimer -
Home Account -
John R Pierce -
Les Mikesell -
m.roth＠5-cent.us -
Negative -
Stephen Harris -
Toby Bluhm