Re: KVM Test result, KVM62-RC2

[Farkas Levente <lfarkas@bppiac.hu>]

Marcelo Tosatti wrote:
> On Tue, Feb 26, 2008 at 02:38:59PM +0100, Farkas Levente wrote:
>> Farkas Levente wrote:
>>> Izik Eidus wrote:
>>>> On Tue, 2008-02-26 at 12:58 +0100, Farkas Levente wrote:
>>>>> Zhao, Yunfeng wrote:
>>>>>> Hi, all,
>>>>>>  
>>>>>> This is today's KVM test result against KVM62-RC2.
>>>>>>
>>>>>> Two old issues fixed:
>>>>>> 1. fc5/fc6/rhel5u1 no-acpi up guests can't boot on pae host
>>>>>> https://sourceforge.net/tracker/index.php?func=detail&aid=1901208&group_
>>>>>> id=180599&atid=893831
>>>>>> 2. Cannot boot 32bit smp RHEL5.1 guest with nic on 64bit host
>>>>>> https://sourceforge.net/tracker/?func=detail&atid=893831&aid=1812043&gro
>>>>>> up_id=180599
>>>>> unfortunately it seems there are other problems with smp on rhel 5.1:-(
>>>>> almost all commands hangs randomly. which means eg:
>>>>> yum update
>>>>> do nothing after 3-4 times and ctrl-c it's runs without problem. the
>>>>> same happened with an rpmbuild command during it's hang nothing happened
>>>>> no cpu load, no error neither in the host no on the guest:
>>>>> -------------------------------
>>>>> 25028 ?        Ss     0:00  \_ sshd: robot [priv]
>>>>> 25030 ?        S      0:02  |   \_ sshd: robot@pts/1
>>>>> 25031 pts/1    Ss     0:00  |       \_ -bash
>>>>> 29543 pts/1    S+     0:00  |           \_ rpmbuild -ta --define ver
>>>>> 4.0.4.15 --define svn 2483 --define dist el5
>>>>> /home/robot/rpm/SOURCES/xyz-2483.tar.bz2
>>>>> -------------------------------
>>>>> so it's finally boot, but very unstable (the same not happened with 64
>>>>> guest on the same host).
>>>> what happen if you run the commands with taskset 0 ?
>>> # taskset 0 ls
>>> sched_setaffinity: Invalid argument
>>> failed to set pid 0's affinity.
>> ok so i've a few more info. my host system has 4 cpu (centos-5.1 x86_64)
>> and if give 4 vcpu to the guest (centos-5.1 i386) it's working correctly
>> IF i do _not_ start any other guest on the host. if i start other
>> guests, then it's randomly start or not start a given command. but with
>> taskset -c 0 it's always runs the command.
>> now i switch back to 1 vcpu since i can't force vcpu (taskset) for cron
>> tasks and other processes:-(
> 
> Farkas,
> 
> You can workaround these problems by using a different, less problematic
> clocksource such as acpipm, until the TSC/migration issues are fully
> resolved.
> 
> Add "clocksource=acpi_pm" to the kernel options.

than just quck question we use the noapic kernel option for the guest (i
don't remember why but there was some problem without it about half year
ago). do i still have to use the noapic option for the kernel? (i hope
somebody remember the reason:-)


-- 
  Levente                               "Si vis pacem para bellum!"

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/