Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 1 Mar 2017 13:50:30 -0500
From:      Allan Jude <allanjude@freebsd.org>
To:        freebsd-virtualization@freebsd.org
Subject:   Re: Timer problems on FreeBSD 11R on KVM
Message-ID:  <4cc78ba2-a23b-5cb2-92c8-6435762e2910@freebsd.org>
In-Reply-To: <20170301174640.GA44226@hades.panopticon>
References:  <20170301174640.GA44226@hades.panopticon>

next in thread | previous in thread | raw e-mail | index | archive | help
This is an OpenPGP/MIME signed message (RFC 4880 and 3156)
--xpQA30Cv1tnuGUkv5JiX3pWTXT1Jv7BJK
Content-Type: multipart/mixed; boundary="JAVBoaMBxLnr6rKUAmCOfePVsDmirRUJJ";
 protected-headers="v1"
From: Allan Jude <allanjude@freebsd.org>
To: freebsd-virtualization@freebsd.org
Message-ID: <4cc78ba2-a23b-5cb2-92c8-6435762e2910@freebsd.org>
Subject: Re: Timer problems on FreeBSD 11R on KVM
References: <20170301174640.GA44226@hades.panopticon>
In-Reply-To: <20170301174640.GA44226@hades.panopticon>

--JAVBoaMBxLnr6rKUAmCOfePVsDmirRUJJ
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

On 2017-03-01 12:46, Dmitry Marakasov wrote:
> Hi!
>=20
> I'm running a bunch of FreeBSD VMs on VPS provider which uses KVM.
> Recently one of them began bahaving strangely. Symptoms:
>=20
> - Much ~(30%) CPU spent in intr
> - top shows CPU used by "intr{swi4: clock (0)}"
> - kernel: calcru: runtime went backwards from 13958 usec to 10792 usec =
for pid 919 (sshd)
>   messages in logs
> - There's no much load, no high network traffic, systat shows no excess=

>   interrupts.
> - This intr load is sporadic in form of ~30 minute busts each several
>   hours
> - The box had ntpd running, but disabling it doesn't help
> - Playing with timer-related sysctls does not help either
>=20
> I've tried i8254 and TSC-low timecounters:
>=20
> kern.timecounter.hardware: ACPI-fast
> kern.timecounter.choice: TSC-low(800) i8254(0) ACPI-fast(900) dummy(-10=
00000)
>=20
> and kern.eventtimer.periodic=3D1
>=20
> to no effect.
>=20
>=20
> I suspect that this began after some changes on the hoster's side.
>=20
> - This VM was working for more than a year without problems before
> - A graph which monitors a reply time of website running on that VM
>   showed min. response time jump from 50 to 75 msec on Monday, after
>   which problems began.
>=20
> So while I'm getting the hoster to tell me what they have changed, any
> ideas?
>=20

Look at:
kern.eventtimer.choice

and try different kern.eventtimer.timer values

--=20
Allan Jude


--JAVBoaMBxLnr6rKUAmCOfePVsDmirRUJJ--

--xpQA30Cv1tnuGUkv5JiX3pWTXT1Jv7BJK
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.16 (MingW32)
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=EFha
-----END PGP SIGNATURE-----

--xpQA30Cv1tnuGUkv5JiX3pWTXT1Jv7BJK--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4cc78ba2-a23b-5cb2-92c8-6435762e2910>