[pve-devel] [PATCH manager] add graphite tcp support

[Martin Verges]

Hello,

yes 1 second should be ok as well and it can be configured. Maybe if
pvestatd has problems with blocking, an upper limit would be good as well
to prevent misconfiguration.

I agree that we don't read from that socket, therefore
"$carbon_socket->read_timeout($timeout);" should not be required. However,
I don't know how perl works internally and therefore I think it won't hurt
to set read and write to prevent any blocking. Maybe someone with better
knowledge about perl and pvestatd should answer that.

--
Martin Verges
Managing director

Mobile: +49 174 9335695
E-Mail: martin.verges at croit.io
Chat: https://t.me/MartinVerges

croit GmbH, Freseniusstr. 31h, 81247 Munich
CEO: Martin Verges - VAT-ID: DE310638492
Com. register: Amtsgericht Munich HRB 231263

Web: https://croit.io
YouTube: https://goo.gl/PGE1Bx


Am Mi., 6. Nov. 2019 um 21:11 Uhr schrieb Thomas Lamprecht <
t.lamprecht at proxmox.com>:

> On 11/5/19 10:34 PM, Martin Verges wrote:
> > This change allows sending statistics to graphite over TCP.
> >
> > So far only UDP is possible, which is not available in some
> environments, like behind a loadbalancer.
> >
> > Configuration example:
> > ~ $ cat /etc/pve/status.cfg
> >
> > graphite:
> >     server 10.20.30.40
> >     port 2003
> >     path proxmox
> >     proto tcp
> >     timeout 3
> >
> > Signed-off-by: Martin Verges <martin.verges at croit.io>
> > ---
> >  PVE/Status/Graphite.pm | 25 ++++++++++++++++++++++++-
> >  1 file changed, 24 insertions(+), 1 deletion(-)
> >
>
> applied, but with a followup reducing the default timeout to 1
> second, which should be still high enough for most people, as TCP
> needs roughly 1.5 time of the Round Trip Time for connection setup,
> So, with 1 second timeout we're still good for connections with 660ms
> latency in-between.
>
> Users can always increase it if required in their environment.
>
> Thanks a lot!
>
>