*Meaning what exactly: block storage like iSCSI for example? *
I think, it is iscsi. It is drive on rhvm virtual machine.
* In that case, it's clearly not a prometheus problem, it's something to do
with your container or your k8s environment. *
ps and top slows only when prometheus use 1 core. On start they are
responsive.
*strace -s 128 -f ps ....etc... 2>strace.out*
strace for ps - hang on
open("/proc/1/cmdline", O_RDONLY) = 3
read(3,
and cat /proc/1/cmdline take a lot of time
strace prometheus process:
strace -p 41850
strace: Process 41850 attached
futex(0x44daec8, FUTEX_WAIT_PRIVATE, 0, NULL
понедельник, 31 августа 2020 г. в 15:39:37 UTC+5, [email protected]:
> On Monday, 31 August 2020 11:31:39 UTC+1, Redneck_prm wrote:
>>
>> disk is some network storage
>>
>
> Meaning what exactly: block storage like iSCSI for example? Prometheus
> does not work well with NFS storage.
>
> I am curious why ps and top slow inside the container? What they need?
>>
>
> In that case, it's clearly not a prometheus problem, it's something to do
> with your container or your k8s environment. Try using strace to see the
> system calls, e.g.
>
> strace -s 128 -f ps ....etc... 2>strace.out
>
> then go through the calls in strace.out. Adding "-ttt" can give you
> better timestamps.
>
> Some possibilities: a problem with DNS; a problem with uid/gid resolution
> in nsswitch (especially if you're using something like an LDAP backend); a
> problem with one of your filesystems.
>
--
You received this message because you are subscribed to the Google Groups
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/prometheus-users/d859c4f5-831f-48c7-b82a-53aabf37baf5n%40googlegroups.com.
