On 08/31/2012 05:43 PM, Osier Yang wrote:
On 2012年08月31日 17:22, Guannan Ren wrote:
> On NUMA machine, the length of string got from file
> cpuacct.usage_percpu is quite large, so expand the
> limit of 1024 bytes.
>
> errors like:
> Failed to read file \
> '/cgroup/cpuacct/libvirt/qemu/rhel6q/cpuacct.usage_percpu': \
> Value too large for defined data type
> ---
> src/util/cgroup.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/util/cgroup.c b/src/util/cgroup.c
> index 8541c7f..5dc0764 100644
> --- a/src/util/cgroup.c
> +++ b/src/util/cgroup.c
> @@ -360,7 +360,7 @@ static int virCgroupGetValueStr(virCgroupPtr group,
>
> VIR_DEBUG("Get value %s", keypath);
>
> - rc = virFileReadAll(keypath, 1024, value);
> + rc = virFileReadAll(keypath, 1024*1024, value);
Can it be that large?
I don't know the good limit for this value.
According to my testing machine that has 40 physical cpus,
160 logical cpus from /proc/cpuinfo.
The size of cpuacct.usage_percpu is 1366