[OpenAFS] afs stalled for large files, opeanfs 1.6.1, ubuntu 12.04, particular network

Liza M lizamijovic@hotmail.com
Wed, 23 Apr 2014 02:14:12 +0200


--_af009fcd-e313-4375-a994-f7ad247a0732_
Content-Type: text/plain; charset="iso-8859-2"
Content-Transfer-Encoding: quoted-printable

=0A=
=0A=
=0A=
Hello Jeff=2C=20

thanks for the feedback and suggestion=2C=20

>=20
> This sounds like an MTU problem -- your connection to the network in
> question includes a segment with a lower MTU than those to which the
> fileserver and client are directly connected.  Perhaps there is a VPN or
> other tunnel in between?

no=2C I am not on VPN and there is no other tunnel between server and clien=
t=20

>=20
> Try running this command (as root) on the client:
>=20
> ip link set dev eth0 mtu 1200
>=20
> Replace "eth0" with the name of the relevant interface.
>=20

relevant interface is eth0 in my case. settining mtu size however does not =
seem to have any effect of files stalling=3B=20

root@lmlpt:/home/lmijovic# ip addr | grep eth0 | head -n 1
2: eth0: <BROADCAST=2CMULTICAST=2CUP=2CLOWER_UP> mtu 1500 qdisc pfifo_fast =
state UP qlen 1000

cp of file to afs stalls at file size of 1329 bytes

root@lmlpt:/home/lmijovic# ip link set dev eth0 mtu 1200
root@lmlpt:/home/lmijovic# ip addr | grep eth0 | head -n 1                 =
                                                                           =
                                  =20
2: eth0: <BROADCAST=2CMULTICAST=2CUP=2CLOWER_UP> mtu 1200 qdisc pfifo_fast =
state UP qlen 1000 =20
+ a couple of different sizes (500=2C1000)

still stalls at wc -c 1329 (while 1328 in all cases gets copied fine/wo sta=
lling)

Best=2C Liza

>=20
> If this makes the problem go away=2C you know that the problem is that
> some portion of the network is not passing packets above the configured
> size.
>=20
>=20
> -- Jeff
>=20
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
=0A=
 		 	   		  =

--_af009fcd-e313-4375-a994-f7ad247a0732_
Content-Type: text/html; charset="iso-8859-2"
Content-Transfer-Encoding: quoted-printable

<html>
<head>
</head>
<body class=3D'hmmessage'><div dir=3D'ltr'>=0A=
=0A=
<style><!--=0A=
.hmmessage P=0A=
{=0A=
margin:0px=3B=0A=
padding:0px=0A=
}=0A=
body.hmmessage=0A=
{=0A=
font-size: 12pt=3B=0A=
font-family:Calibri=0A=
}=0A=
--></style>=0A=
<div dir=3D"ltr">Hello Jeff=2C <br><br>thanks for the feedback and suggesti=
on=2C <br><br>&gt=3B <br><div>&gt=3B This sounds like an MTU problem -- you=
r connection to the network in<br>&gt=3B question includes a segment with a=
 lower MTU than those to which the<br>&gt=3B fileserver and client are dire=
ctly connected.  Perhaps there is a VPN or<br>&gt=3B other tunnel in betwee=
n?<br><br>no=2C I am not on VPN and there is no other tunnel between server=
 and client <br><br>&gt=3B <br>&gt=3B Try running this command (as root) on=
 the client:<br>&gt=3B <br>&gt=3B ip link set dev eth0 mtu 1200<br>&gt=3B <=
br>&gt=3B Replace "eth0" with the name of the relevant interface.<br>&gt=3B=
 <br><br>relevant interface is eth0 in my case. settining mtu size however =
does not seem to have any effect of files stalling=3B <br><br>root@lmlpt:/h=
ome/lmijovic# ip addr | grep eth0 | head -n 1<br>2: eth0: &lt=3BBROADCAST=
=2CMULTICAST=2CUP=2CLOWER_UP&gt=3B mtu 1500 qdisc pfifo_fast state UP qlen =
1000<br><br>cp of file to afs stalls at file size of 1329 bytes<br><br>root=
@lmlpt:/home/lmijovic# ip link set dev eth0 mtu 1200<br>root@lmlpt:/home/lm=
ijovic# ip addr | grep eth0 | head -n 1&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbs=
p=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=
=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B=
&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nb=
sp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=
=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B=
&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nb=
sp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=
=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B=
&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nb=
sp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=
=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B=
&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nb=
sp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=3B&nbsp=
=3B <br>2: eth0: &lt=3BBROADCAST=2CMULTICAST=2CUP=2CLOWER_UP&gt=3B mtu 1200=
 qdisc pfifo_fast state UP qlen 1000&nbsp=3B <br>+ a couple of different si=
zes (500=2C1000)<br><br>still stalls at wc -c 1329 (while 1328 in all cases=
 gets copied fine/wo stalling)<br><br>Best=2C Liza<br><br>&gt=3B <br>&gt=3B=
 If this makes the problem go away=2C you know that the problem is that<br>=
&gt=3B some portion of the network is not passing packets above the configu=
red<br>&gt=3B size.<br>&gt=3B <br>&gt=3B <br>&gt=3B -- Jeff<br>&gt=3B <br>&=
gt=3B _______________________________________________<br>&gt=3B OpenAFS-inf=
o mailing list<br>&gt=3B OpenAFS-info@openafs.org<br>&gt=3B https://lists.o=
penafs.org/mailman/listinfo/openafs-info<br></div></div>=0A=
 		 	   		  </div></body>
</html>=

--_af009fcd-e313-4375-a994-f7ad247a0732_--