Re: Clock jumps
On 05/26/2010 11:31 AM, Alexander Graf wrote: On 26.05.2010, at 19:10, Orion Poplawski wrote: On 05/25/2010 12:21 AM, Gleb Natapov wrote: Adding kvm to CC. On Mon, May 24, 2010 at 04:06:32PM +, Orion Poplawski wrote: I have a KVM virtual machine running 2.6.33.4-95.fc13.x86_64 on a CentOS 5.5 host whose clock jumps about 8-12 hours a couple times a day. I have no idea what is causing it. Fedora 12 and Centos 5.5 KVM machines run fine on the same host. Is there any debugging I can enable to see what is jumping the clock? kvm-clock: cpu 0, msr 0:1ba4741, boot clock kvm-clock: cpu 0, msr 0:1e15741, primary cpu clock Switching to clocksource kvm-clock rtc_cmos 00:01: setting system clock to 2010-05-20 16:59:48 UTC (1274374788) Thanks, though I don't think it made it there. I'm also not sure it's completely limited to KVM, though that is the only running system I am currently seeing the problem on. I also see clock jumps during anaconda installs on physical hardware and apparently they have been present since at least F11. Might be unrelated though. I'm really at a loss of how to debug this though. Do you have ntpd running inside the guest? I have a bug report lying around about 2.6.33 with kvm-clock jumping in time when ntpd is used: https://bugzilla.novell.com/show_bug.cgi?id=582260 Alex Turning off ntpd and chronyd did not help for me. -- Orion Poplawski Technical Manager 303-415-9701 x222 NWRA/CoRA DivisionFAX: 303-415-9702 3380 Mitchell Lane or...@cora.nwra.com Boulder, CO 80301 http://www.cora.nwra.com -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
On Fri, 2010-05-28 at 02:33 +0200, Bernhard Schmidt wrote: > On 28.05.2010 02:00, john stultz wrote: > > Looking at the diff: > > --- dmesg-lenny 2010-05-27 16:45:33.0 -0700 > > +++ dmesg-squeeze 2010-05-27 16:46:14.0 -0700 > > @@ -132,8 +132,8 @@ > > console [ttyS1] enabled > > hpet clockevent registered > > Fast TSC calibration using PIT > > -Detected 2660.398 MHz processor. > > -Calibrating delay loop (skipped), value calculated using timer frequency.. > > 5320.79 BogoMIPS (lpj=10641592) > > +Detected 2613.324 MHz processor. > > +Calibrating delay loop (skipped), value calculated using timer frequency.. > > 5226.64 BogoMIPS (lpj=10453296) > > Security Framework initialized > > Dentry cache hash table entries: 524288 (order: 10, 4194304 bytes) > > Inode-cache hash table entries: 262144 (order: 9, 2097152 bytes) > > @@ -160,7 +160,7 @@ > > CPU0: Intel(R) Xeon(R) CPU3075 @ 2.66GHz stepping 0b > > Booting Node 0, Processors #1 > > Brought up 2 CPUs > > -Total of 2 processors activated (10640.79 BogoMIPS). > > +Total of 2 processors activated (10546.63 BogoMIPS). > > NET: Registered protocol family 16 > > ACPI: bus type pci registered > > PCI: MMCONFIG for domain [bus 00-ff] at [mem 0xe000-0xefff] > > (base 0xe000) > > > > So you can see in the above the during the second boot the TSC > > calibration was badly mis-calculated. This was the cause of the skew. > > > > Not sure how that might be linked to the distro upgrade. It could have > > been something like SMI damage during the calibration time, but I > > thought the calibration loop watched for that. > > > > Bernhard: I expect with all those vms, this machine isn't rebooted > > frequently. So could you look at the logs to see how much the "Detected > > .yyy MHz processor." line varies by across a few other boots (if > > they still exist?). > > Correct, the box isn't rebooted often, but I do have a few dmesg outputs > laying around. lpj was always almost the same until the very last boot > which screwed up the clock. > > dmesg:[0.00] Linux version 2.6.33 (r...@svr02) (gcc version > 4.3.2 (Debian 4.3.2-1.1) ) #1 SMP Sun Mar 7 23:01:45 CET 2010 > dmesg:[0.008005] Calibrating delay loop (skipped), value calculated > using timer frequency.. 5226.64 BogoMIPS (lpj=10453296) > dmesg:[0.288002] Total of 2 processors activated (10546.63 BogoMIPS). Yea. The bogomips/loops per jiffies are actually calculated with a different chunk of code (although its interesting it miscalculated in both cases). Could you send the "Detected .yyy MHz processor." lines as well? thanks -john -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
On 28.05.2010 02:00, john stultz wrote: Hi John, Looking at the diff: --- dmesg-lenny 2010-05-27 16:45:33.0 -0700 +++ dmesg-squeeze 2010-05-27 16:46:14.0 -0700 @@ -132,8 +132,8 @@ console [ttyS1] enabled hpet clockevent registered Fast TSC calibration using PIT -Detected 2660.398 MHz processor. -Calibrating delay loop (skipped), value calculated using timer frequency.. 5320.79 BogoMIPS (lpj=10641592) +Detected 2613.324 MHz processor. +Calibrating delay loop (skipped), value calculated using timer frequency.. 5226.64 BogoMIPS (lpj=10453296) Security Framework initialized Dentry cache hash table entries: 524288 (order: 10, 4194304 bytes) Inode-cache hash table entries: 262144 (order: 9, 2097152 bytes) @@ -160,7 +160,7 @@ CPU0: Intel(R) Xeon(R) CPU3075 @ 2.66GHz stepping 0b Booting Node 0, Processors #1 Brought up 2 CPUs -Total of 2 processors activated (10640.79 BogoMIPS). +Total of 2 processors activated (10546.63 BogoMIPS). NET: Registered protocol family 16 ACPI: bus type pci registered PCI: MMCONFIG for domain [bus 00-ff] at [mem 0xe000-0xefff] (base 0xe000) So you can see in the above the during the second boot the TSC calibration was badly mis-calculated. This was the cause of the skew. Not sure how that might be linked to the distro upgrade. It could have been something like SMI damage during the calibration time, but I thought the calibration loop watched for that. Bernhard: I expect with all those vms, this machine isn't rebooted frequently. So could you look at the logs to see how much the "Detected .yyy MHz processor." line varies by across a few other boots (if they still exist?). Correct, the box isn't rebooted often, but I do have a few dmesg outputs laying around. lpj was always almost the same until the very last boot which screwed up the clock. dmesg:[0.00] Linux version 2.6.33 (r...@svr02) (gcc version 4.3.2 (Debian 4.3.2-1.1) ) #1 SMP Sun Mar 7 23:01:45 CET 2010 dmesg:[0.008005] Calibrating delay loop (skipped), value calculated using timer frequency.. 5226.64 BogoMIPS (lpj=10453296) dmesg:[0.288002] Total of 2 processors activated (10546.63 BogoMIPS). dmesg.0:[0.00] Linux version 2.6.33 (r...@svr02) (gcc version 4.3.2 (Debian 4.3.2-1.1) ) #1 SMP Sun Mar 7 23:01:45 CET 2010 dmesg.0:[0.008005] Calibrating delay loop (skipped), value calculated using timer frequency.. 5320.79 BogoMIPS (lpj=10641592) dmesg.0:[0.274022] Total of 2 processors activated (10640.79 BogoMIPS). dmesg.1.gz:[0.00] Linux version 2.6.32-rc7 (r...@svr02) (gcc version 4.3.2 (Debian 4.3.2-1.1) ) #1 SMP Thu Nov 19 14:36:03 CET 2009 dmesg.1.gz:[0.012004] Calibrating delay loop (skipped), value calculated using timer frequency.. 5319.06 BogoMIPS (lpj=10638120) dmesg.1.gz:[0.016000] Calibrating delay using timer specific routine.. 5319.99 BogoMIPS (lpj=10639980) dmesg.1.gz:[0.260003] Total of 2 processors activated (10639.05 BogoMIPS). dmesg.2.gz:[0.00] Linux version 2.6.32-rc7 (r...@svr02) (gcc version 4.3.2 (Debian 4.3.2-1.1) ) #1 SMP Thu Nov 19 14:36:03 CET 2009 dmesg.2.gz:[0.012005] Calibrating delay loop (skipped), value calculated using timer frequency.. 5319.35 BogoMIPS (lpj=10638712) dmesg.2.gz:[0.016000] Calibrating delay using timer specific routine.. 5319.99 BogoMIPS (lpj=10639990) dmesg.2.gz:[0.261567] Total of 2 processors activated (10639.35 BogoMIPS). dmesg.3.gz:[0.00] Linux version 2.6.32-rc7 (r...@svr02) (gcc version 4.3.2 (Debian 4.3.2-1.1) ) #1 SMP Thu Nov 19 14:36:03 CET 2009 dmesg.3.gz:[0.012005] Calibrating delay loop (skipped), value calculated using timer frequency.. 5319.97 BogoMIPS (lpj=10639956) dmesg.3.gz:[0.016000] Calibrating delay using timer specific routine.. 5319.99 BogoMIPS (lpj=10639987) dmesg.3.gz:[0.257152] Total of 2 processors activated (10639.97 BogoMIPS). dmesg.4.gz:[0.00] Linux version 2.6.32-rc7 (r...@svr02) (gcc version 4.3.2 (Debian 4.3.2-1.1) ) #1 SMP Thu Nov 19 14:36:03 CET 2009 dmesg.4.gz:[0.012005] Calibrating delay loop (skipped), value calculated using timer frequency.. 5319.84 BogoMIPS (lpj=10639688) dmesg.4.gz:[0.016000] Calibrating delay using timer specific routine.. 5319.99 BogoMIPS (lpj=10639993) dmesg.4.gz:[0.253571] Total of 2 processors activated (10639.84 BogoMIPS). If necessary I can reboot once more, but I'd like to avoid it. Bernhard -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
On Thu, 2010-05-27 at 23:48 +0200, Bernhard Schmidt wrote: > On 27.05.2010 21:08, john stultz wrote: > > I'd be very interested in hearing more about the host side issue. So > > this happened with the same kernel that you were using before, with no > > trouble? > > Correct. > > > Could you also send dmesg output from this boot? And if you can find > > any older dmesg logs to compare with, send those too? > > See http://users.birkenwald.de/~berni/temp/dmesg-lenny and > http://users.birkenwald.de/~berni/temp/dmesg-squeeze . Although running > on the same kernel binary the initrd changed greatly when upgrading, so > ordering/timing between those two is off. > > Note that the dmesg output is captured right after boot. I think I > remember seeing a "TSC unstable" message pretty soon after boot, but I > might be mixing it up with my other AMD-based KVM server. I don't hold > normal (non-boot) logs that long, so I can't tell for sure. Looking at the diff: --- dmesg-lenny 2010-05-27 16:45:33.0 -0700 +++ dmesg-squeeze 2010-05-27 16:46:14.0 -0700 @@ -132,8 +132,8 @@ console [ttyS1] enabled hpet clockevent registered Fast TSC calibration using PIT -Detected 2660.398 MHz processor. -Calibrating delay loop (skipped), value calculated using timer frequency.. 5320.79 BogoMIPS (lpj=10641592) +Detected 2613.324 MHz processor. +Calibrating delay loop (skipped), value calculated using timer frequency.. 5226.64 BogoMIPS (lpj=10453296) Security Framework initialized Dentry cache hash table entries: 524288 (order: 10, 4194304 bytes) Inode-cache hash table entries: 262144 (order: 9, 2097152 bytes) @@ -160,7 +160,7 @@ CPU0: Intel(R) Xeon(R) CPU3075 @ 2.66GHz stepping 0b Booting Node 0, Processors #1 Brought up 2 CPUs -Total of 2 processors activated (10640.79 BogoMIPS). +Total of 2 processors activated (10546.63 BogoMIPS). NET: Registered protocol family 16 ACPI: bus type pci registered PCI: MMCONFIG for domain [bus 00-ff] at [mem 0xe000-0xefff] (base 0xe000) So you can see in the above the during the second boot the TSC calibration was badly mis-calculated. This was the cause of the skew. Not sure how that might be linked to the distro upgrade. It could have been something like SMI damage during the calibration time, but I thought the calibration loop watched for that. Bernhard: I expect with all those vms, this machine isn't rebooted frequently. So could you look at the logs to see how much the "Detected .yyy MHz processor." line varies by across a few other boots (if they still exist?). Ingo/Thomas: Any thoughts, should we be considering dropping the quick_pit_calibrate() code and always do the slower route? thanks -john -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
On 05/27/2010 12:12 PM, Bernhard Schmidt wrote: On 27.05.2010 23:53, Zachary Amsden wrote: Hello Zachary, I have server side fixes for this kvm-clock which seem to give me a stable clock on this machine, but for true SMP stability, you will need Glauber's guest side changes to kvmclock as well. It is impossible to guarantee strictly monotonic clocksource across multiple CPUs when frequency is dynamically changing (and also because of the C1E idle problems). Is all this relevant only when the host is on TSC? Because I have seen these jumps when the host was on HPET and the guests were using kvm-clock. It doesn't matter what the host uses (although the host on TSC with unstable TSC can make things worse), tsc and kvmclock sources in the guest will be unstable regardless. -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
On 05/27/2010 12:12 PM, Bernhard Schmidt wrote: On 27.05.2010 23:53, Zachary Amsden wrote: Hello Zachary, I have server side fixes for this kvm-clock which seem to give me a stable clock on this machine, but for true SMP stability, you will need Glauber's guest side changes to kvmclock as well. It is impossible to guarantee strictly monotonic clocksource across multiple CPUs when frequency is dynamically changing (and also because of the C1E idle problems). Is all this relevant only when the host is on TSC? Because I have seen these jumps when the host was on HPET and the guests were using kvm-clock. Anyway, can you send me both patches? I'd like to try it, but I have completely lost track where the up-to-date patches are. There's more than two, there's quite a bit, I'll send it to the list soon. Zach -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
On 27.05.2010 23:53, Zachary Amsden wrote: Hello Zachary, I have server side fixes for this kvm-clock which seem to give me a stable clock on this machine, but for true SMP stability, you will need Glauber's guest side changes to kvmclock as well. It is impossible to guarantee strictly monotonic clocksource across multiple CPUs when frequency is dynamically changing (and also because of the C1E idle problems). Is all this relevant only when the host is on TSC? Because I have seen these jumps when the host was on HPET and the guests were using kvm-clock. Anyway, can you send me both patches? I'd like to try it, but I have completely lost track where the up-to-date patches are. Bernhard -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
On 27.05.2010 21:08, john stultz wrote: Hi John, I'd be very interested in hearing more about the host side issue. So this happened with the same kernel that you were using before, with no trouble? Correct. Could you also send dmesg output from this boot? And if you can find any older dmesg logs to compare with, send those too? See http://users.birkenwald.de/~berni/temp/dmesg-lenny and http://users.birkenwald.de/~berni/temp/dmesg-squeeze . Although running on the same kernel binary the initrd changed greatly when upgrading, so ordering/timing between those two is off. Note that the dmesg output is captured right after boot. I think I remember seeing a "TSC unstable" message pretty soon after boot, but I might be mixing it up with my other AMD-based KVM server. I don't hold normal (non-boot) logs that long, so I can't tell for sure. If you need any more info feel free to contact me. Bernhard -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
On 05/27/2010 08:32 AM, Bernhard Schmidt wrote: Alexander Graf wrote: Hi, Do you have ntpd running inside the guest? I have a bug report lying around about 2.6.33 with kvm-clock jumping in time when ntpd is used: https://bugzilla.novell.com/show_bug.cgi?id=582260 I want to chime in here, I have a very similar problem, but not with ntpd in the guest. The host was a HP ProLiant DL320 G5p with a Dualcore Xeon3075. System was a Debian Lenny with a custom 2.6.33 host kernel and a custom qemu-kvm 0.11.0 .deb ported from Ubuntu. The host is synced with ntpd. The guests are various Debian Lenny/Squeeze VMs, with a custom kernel (2.6.33 at the moment) with kvm-clock. Exclusively amd64 guest kernels, but one system has i386 userland. With this setup once in a while (maybe every other week) one VM would have a sudden clock jump, 6-12 hours into the future. No kernel messages or other log entries than applications complaining about the clock jump after the fact. Other VMs were unaffected. Yesterday I did an upgrade to Debian Squeeze. This involved a new qemu-kvm (0.12.4), but not a new host kernel. I also upgraded the guest kernels from 2.6.33 to 2.6.33.4. First of all, after the reboot the host clock was totally unreliable. I had a constant skew of up to five seconds per minute in the host clock, which of course affected the VMs as well. This problem went away when I changed from tsc into hpet on the host. The host does CPU frequency scaling which is, as far as I know, known to affect TSC stability. I think I remember messages about tsc being unstable in earlier boots, maybe the detection did just not work this time. Worse, the clock jump issues in the guest appeared much more often than before. The higher loaded VMs did not survive ten minutes without jumping several hours ahead. Situation has stabilized after setting clocksource hpet in the guest immediately after boot. So it seems kvm-clock has some issues here. I've seen a preliminary patch floating around on the ML by Zachary Amsden, but I haven't tried it yet. It talks of backward warps, but so far I've only seen forward warps (the VM time suddenly jumps into the future), so it might be unrelated. I have an AMD Turion TL-52 machine with unreliable TSC. It varies with CPU frequency, which is okay, we can compensate for that, but worse, it has broken clocking when in C1E idle. Apparently, it divides down the clock input to an idle core, so it only runs at 1/16 or whatever of the rate, and adds a multiplier to the TSC increment, so it scales by 16 instead of by 1 (whatever the actual numbers are I don't know, but this illustrates the point). When it wakes up to service a cache probe from another core, it now runs with full clock rate ... and still uses the multiplier for the TSC increment. The effect is that idle CPUs have TSC which may increase faster than running CPUs. Given time, this delta can add to a very large number (in theory, it's a random walk, but it can go very far off). If a VM is running on this CPU and happens to match the idle pattern without switching CPUs, time can effectively run accelerated on that VM, and very rapidly things are going to get confused. Newer kernels should detect the host clock being unreliable quite quickly; my F13 machine detects it right away at boot. I have server side fixes for this kvm-clock which seem to give me a stable clock on this machine, but for true SMP stability, you will need Glauber's guest side changes to kvmclock as well. It is impossible to guarantee strictly monotonic clocksource across multiple CPUs when frequency is dynamically changing (and also because of the C1E idle problems). There is one remaining problem to fix, the reset of TSC on reboot in SMP will destabilize the TSCs again, but now I've actually got VMs running again (different bug), that shouldn't be long. Zach -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
On Thu, May 27, 2010 at 11:32 AM, Bernhard Schmidt wrote: > Alexander Graf wrote: >> Do you have ntpd running inside the guest? I have a bug report lying >> around about 2.6.33 with kvm-clock jumping in time when ntpd is used: >> https://bugzilla.novell.com/show_bug.cgi?id=582260 > > I want to chime in here, I have a very similar problem, but not with > ntpd in the guest. > > The host was a HP ProLiant DL320 G5p with a Dualcore Xeon3075. System > was a Debian Lenny with a custom 2.6.33 host kernel and a custom > qemu-kvm 0.11.0 .deb ported from Ubuntu. The host is synced with ntpd. > > The guests are various Debian Lenny/Squeeze VMs, with a custom kernel > (2.6.33 at the moment) with kvm-clock. Exclusively amd64 guest > kernels, but one system has i386 userland. > > With this setup once in a while (maybe every other week) one VM would > have a sudden clock jump, 6-12 hours into the future. No kernel messages > or other log entries than applications complaining about the clock jump > after the fact. Other VMs were unaffected. > > Yesterday I did an upgrade to Debian Squeeze. This involved a new > qemu-kvm (0.12.4), but not a new host kernel. I also upgraded the guest > kernels from 2.6.33 to 2.6.33.4. > > First of all, after the reboot the host clock was totally unreliable. I > had a constant skew of up to five seconds per minute in the host clock, > which of course affected the VMs as well. This problem went away when I > changed from tsc into hpet on the host. The host does CPU frequency > scaling which is, as far as I know, known to affect TSC stability. I > think I remember messages about tsc being unstable in earlier boots, > maybe the detection did just not work this time. I'd be very interested in hearing more about the host side issue. So this happened with the same kernel that you were using before, with no trouble? Could you also send dmesg output from this boot? And if you can find any older dmesg logs to compare with, send those too? thanks -john -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
Alexander Graf wrote: Hi, > Do you have ntpd running inside the guest? I have a bug report lying > around about 2.6.33 with kvm-clock jumping in time when ntpd is used: > https://bugzilla.novell.com/show_bug.cgi?id=582260 I want to chime in here, I have a very similar problem, but not with ntpd in the guest. The host was a HP ProLiant DL320 G5p with a Dualcore Xeon3075. System was a Debian Lenny with a custom 2.6.33 host kernel and a custom qemu-kvm 0.11.0 .deb ported from Ubuntu. The host is synced with ntpd. The guests are various Debian Lenny/Squeeze VMs, with a custom kernel (2.6.33 at the moment) with kvm-clock. Exclusively amd64 guest kernels, but one system has i386 userland. With this setup once in a while (maybe every other week) one VM would have a sudden clock jump, 6-12 hours into the future. No kernel messages or other log entries than applications complaining about the clock jump after the fact. Other VMs were unaffected. Yesterday I did an upgrade to Debian Squeeze. This involved a new qemu-kvm (0.12.4), but not a new host kernel. I also upgraded the guest kernels from 2.6.33 to 2.6.33.4. First of all, after the reboot the host clock was totally unreliable. I had a constant skew of up to five seconds per minute in the host clock, which of course affected the VMs as well. This problem went away when I changed from tsc into hpet on the host. The host does CPU frequency scaling which is, as far as I know, known to affect TSC stability. I think I remember messages about tsc being unstable in earlier boots, maybe the detection did just not work this time. Worse, the clock jump issues in the guest appeared much more often than before. The higher loaded VMs did not survive ten minutes without jumping several hours ahead. Situation has stabilized after setting clocksource hpet in the guest immediately after boot. So it seems kvm-clock has some issues here. I've seen a preliminary patch floating around on the ML by Zachary Amsden, but I haven't tried it yet. It talks of backward warps, but so far I've only seen forward warps (the VM time suddenly jumps into the future), so it might be unrelated. Bernhard -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
On 05/26/2010 11:31 AM, Alexander Graf wrote: On 26.05.2010, at 19:10, Orion Poplawski wrote: On 05/25/2010 12:21 AM, Gleb Natapov wrote: Adding kvm to CC. On Mon, May 24, 2010 at 04:06:32PM +, Orion Poplawski wrote: I have a KVM virtual machine running 2.6.33.4-95.fc13.x86_64 on a CentOS 5.5 host whose clock jumps about 8-12 hours a couple times a day. I have no idea what is causing it. Fedora 12 and Centos 5.5 KVM machines run fine on the same host. Is there any debugging I can enable to see what is jumping the clock? kvm-clock: cpu 0, msr 0:1ba4741, boot clock kvm-clock: cpu 0, msr 0:1e15741, primary cpu clock Switching to clocksource kvm-clock rtc_cmos 00:01: setting system clock to 2010-05-20 16:59:48 UTC (1274374788) Thanks, though I don't think it made it there. I'm also not sure it's completely limited to KVM, though that is the only running system I am currently seeing the problem on. I also see clock jumps during anaconda installs on physical hardware and apparently they have been present since at least F11. Might be unrelated though. I'm really at a loss of how to debug this though. Do you have ntpd running inside the guest? I have a bug report lying around about 2.6.33 with kvm-clock jumping in time when ntpd is used: https://bugzilla.novell.com/show_bug.cgi?id=582260 Alex That bug looks just like what I'm seeing. I even see the soft lockup messages sometimes as well. May actually be seeing it with a Fedora 12 guest as well - but it results in a hard hang. -- Orion Poplawski Technical Manager 303-415-9701 x222 NWRA/CoRA DivisionFAX: 303-415-9702 3380 Mitchell Lane or...@cora.nwra.com Boulder, CO 80301 http://www.cora.nwra.com -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
On 05/26/2010 11:31 AM, Alexander Graf wrote: Do you have ntpd running inside the guest? I have a bug report lying around about 2.6.33 with kvm-clock jumping in time when ntpd is used: https://bugzilla.novell.com/show_bug.cgi?id=582260 Alex I've used ntpd and chronyd. I haven't tried running without either. I'll do that now... -- Orion Poplawski Technical Manager 303-415-9701 x222 NWRA/CoRA DivisionFAX: 303-415-9702 3380 Mitchell Lane or...@cora.nwra.com Boulder, CO 80301 http://www.cora.nwra.com -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
On 26.05.2010, at 19:10, Orion Poplawski wrote: > On 05/25/2010 12:21 AM, Gleb Natapov wrote: >> Adding kvm to CC. >> >> On Mon, May 24, 2010 at 04:06:32PM +, Orion Poplawski wrote: >>> I have a KVM virtual machine running 2.6.33.4-95.fc13.x86_64 on a CentOS 5.5 >>> host whose clock jumps about 8-12 hours a couple times a day. I have no >>> idea >>> what is causing it. Fedora 12 and Centos 5.5 KVM machines run fine on the >>> same >>> host. Is there any debugging I can enable to see what is jumping the clock? >>> >>> kvm-clock: cpu 0, msr 0:1ba4741, boot clock >>> kvm-clock: cpu 0, msr 0:1e15741, primary cpu clock >>> Switching to clocksource kvm-clock >>> rtc_cmos 00:01: setting system clock to 2010-05-20 16:59:48 UTC (1274374788) > > > Thanks, though I don't think it made it there. I'm also not sure it's > completely limited to KVM, though that is the only running system I am > currently seeing the problem on. I also see clock jumps during anaconda > installs on physical hardware and apparently they have been present since at > least F11. Might be unrelated though. > > I'm really at a loss of how to debug this though. Do you have ntpd running inside the guest? I have a bug report lying around about 2.6.33 with kvm-clock jumping in time when ntpd is used: https://bugzilla.novell.com/show_bug.cgi?id=582260 Alex -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
On 05/25/2010 12:21 AM, Gleb Natapov wrote: Adding kvm to CC. On Mon, May 24, 2010 at 04:06:32PM +, Orion Poplawski wrote: I have a KVM virtual machine running 2.6.33.4-95.fc13.x86_64 on a CentOS 5.5 host whose clock jumps about 8-12 hours a couple times a day. I have no idea what is causing it. Fedora 12 and Centos 5.5 KVM machines run fine on the same host. Is there any debugging I can enable to see what is jumping the clock? kvm-clock: cpu 0, msr 0:1ba4741, boot clock kvm-clock: cpu 0, msr 0:1e15741, primary cpu clock Switching to clocksource kvm-clock rtc_cmos 00:01: setting system clock to 2010-05-20 16:59:48 UTC (1274374788) Thanks, though I don't think it made it there. I'm also not sure it's completely limited to KVM, though that is the only running system I am currently seeing the problem on. I also see clock jumps during anaconda installs on physical hardware and apparently they have been present since at least F11. Might be unrelated though. I'm really at a loss of how to debug this though. -- Orion Poplawski Technical Manager 303-415-9701 x222 NWRA/CoRA DivisionFAX: 303-415-9702 3380 Mitchell Lane or...@cora.nwra.com Boulder, CO 80301 http://www.cora.nwra.com -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Clock jumps
Adding kvm to CC. On Mon, May 24, 2010 at 04:06:32PM +, Orion Poplawski wrote: > I have a KVM virtual machine running 2.6.33.4-95.fc13.x86_64 on a CentOS 5.5 > host whose clock jumps about 8-12 hours a couple times a day. I have no idea > what is causing it. Fedora 12 and Centos 5.5 KVM machines run fine on the > same > host. Is there any debugging I can enable to see what is jumping the clock? > > kvm-clock: cpu 0, msr 0:1ba4741, boot clock > kvm-clock: cpu 0, msr 0:1e15741, primary cpu clock > Switching to clocksource kvm-clock > rtc_cmos 00:01: setting system clock to 2010-05-20 16:59:48 UTC (1274374788) > > Thanks, > > Orion > > -- > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to majord...@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > Please read the FAQ at http://www.tux.org/lkml/ -- Gleb. -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html