[OpenAFS] Re: vos move fails to work on new openafs fileserver, seems fine from
other hosts
Fred Drueck
fdruec1@uic.edu
Wed, 11 Apr 2018 18:16:47 -0500
--001a1149a0ea4c204305699ad5ec
Content-Type: text/plain; charset="UTF-8"
I have found the answer to my own problem; the issue was a line in the:
/etc/hosts
file like:
127.0.1.1 newserver.math.uic.edu newserver
So the AFS server was trying to talk to it's loopback address while
external hosts were connecting to the external IP address.
Mystery solved!
-Fred
On Thu, Apr 5, 2018 at 7:15 PM, Fred Drueck <fdruec1@uic.edu> wrote:
>
>
> On Thu, Apr 5, 2018 at 6:56 PM, Fred Drueck <fdruec1@uic.edu> wrote:
>
>> Hello other OpenAFS users,
>>
>> I've recently setup a new file server to try and facilitate upgrading our
>> OpenAFS servers to a newer version of debian (the current stable distro:
>> stretch).
>>
>> The server and db server setup seems to have gone smoothly, but I've got
>> problems with moving the AFS volumes.
>>
>> If I try to do a vos move from the old fileserver to the new one, the
>> move will fail with:
>>
>> Failed to move data for the volume 536881885
>>
>> VOLSER: Problems encountered in doing the dump !
>>
>> However, if I work on the old file server, or another openafs client
>> system (in fact, I've got another debian stretch system using the exact
>> same version the debian openafs-client package) I can move the volume to
>> the new server.
>>
>> And I can verify by looking at the /vicepc directory that the file is
>> there, the vldb says it's there but if I try to move the file back to the
>> older server working on the new server it will fail with a message that the
>> volume is not found, something like:
>>
>> The volume 536881885 is not on the specified site.
>> The current site is : server newserver.math.uic.edu partition /vicepc
>>
>> OK ... but that's exactly the arguments I passed to vos as the
>> -fromserver and -toserver
>>
>> And, if I switch over to another openafs-client or the old AFS file
>> server I can issue the vos command to move the volume back just fine.
>>
>> Anybody have any ideas what's going wrong?
>>
>> -Fred
>>
>
> Looking through the logs, the only thing I saw that looked maybe relevant
> was:
>
> Thu Apr 5 19:02:45 2018 VReadVolumeDiskHeader: Couldn't open header for
> volume 536881887 (errno 2).
>
> but if that's such a problem then why can the other AFS clients happily
> manipulate the volume?
>
> It's puzzling.
>
> -Fred
>
--001a1149a0ea4c204305699ad5ec
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
<div dir=3D"ltr">I have found the answer to my own problem; the issue was a=
line in the:<div><br></div><div>/etc/hosts</div><div><br></div><div>file l=
ike:</div><div><br></div><div>127.0.1.1<span style=3D"white-space:pre"> </s=
pan><a href=3D"http://newserver.math.uic.edu">newserver.math.uic.edu</a><sp=
an style=3D"white-space:pre"> </span>newserver<br></div><div><br></div><div=
>So the AFS server was trying to talk to it's loopback address while ex=
ternal hosts were connecting to the external IP address.</div><div><br></di=
v><div>Mystery solved!</div><div><br></div><div>-Fred</div></div><div class=
=3D"gmail_extra"><br><div class=3D"gmail_quote">On Thu, Apr 5, 2018 at 7:15=
PM, Fred Drueck <span dir=3D"ltr"><<a href=3D"mailto:fdruec1@uic.edu" t=
arget=3D"_blank">fdruec1@uic.edu</a>></span> wrote:<br><blockquote class=
=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padd=
ing-left:1ex"><div dir=3D"ltr"><div><div class=3D"h5"><div><br></div><div c=
lass=3D"gmail_extra"><br><div class=3D"gmail_quote">On Thu, Apr 5, 2018 at =
6:56 PM, Fred Drueck <span dir=3D"ltr"><<a href=3D"mailto:fdruec1@uic.ed=
u" target=3D"_blank">fdruec1@uic.edu</a>></span> wrote:<br><blockquote c=
lass=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px soli=
d rgb(204,204,204);padding-left:1ex"><div dir=3D"ltr">Hello other OpenAFS u=
sers,<div><br></div><div>I've recently setup a new file server to try a=
nd facilitate upgrading our OpenAFS servers to a newer version of debian (t=
he current stable distro: stretch).</div><div><br></div><div>The server and=
db server setup seems to have gone smoothly, but I've got problems wit=
h moving the AFS volumes.</div><div><br></div><div>If I try to do a vos mov=
e from the old fileserver to the new one, the move will fail with:</div><di=
v><br></div><div><div>Failed to move data for the volume 536881885=C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0</div><div>=C2=A0 =C2=A0VOLSER: Problems encountered in doing =
the dump !</div></div><div><br></div><div>However, if I work on the old fil=
e server, or another openafs client system (in fact, I've got another d=
ebian stretch system using the exact same version the debian openafs-client=
package) I can move the volume to the new server.</div><div><br></div><div=
>And I can verify by looking at the /vicepc directory that the file is ther=
e, the vldb says it's there but if I try to move the file back to the o=
lder server working on the new server it will fail with a message that the =
volume is not found, something like:</div><div><br></div><div><div>The volu=
me 536881885 is not on the specified site.</div><div>The current site is : =
server <a href=3D"http://newserver.math.uic.edu" target=3D"_blank">newserve=
r.math.uic.edu</a> partition /vicepc</div></div><div><br></div><div>OK ... =
but that's exactly the arguments I passed to vos as the -fromserver and=
-toserver</div><div><br></div><div>And, if I switch over to another openaf=
s-client or the old AFS file server I can issue the vos command to move the=
volume back just fine.</div><div><br></div><div>Anybody have any ideas wha=
t's going wrong?</div><span class=3D"m_2743398588055249824gmail-HOEnZb"=
><font color=3D"#888888"><div><br></div><div>-Fred</div></font></span></div=
>
</blockquote></div><br></div></div></div><div class=3D"gmail_extra"><span s=
tyle=3D"color:rgb(34,34,34);font-family:arial,sans-serif;font-size:small;fo=
nt-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font=
-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-tra=
nsform:none;white-space:normal;word-spacing:0px;background-color:rgb(255,25=
5,255);text-decoration-style:initial;text-decoration-color:initial;float:no=
ne;display:inline">Looking through the logs, the only thing I saw that look=
ed maybe relevant was:</span><br class=3D"m_2743398588055249824gmail-Apple-=
interchange-newline"></div><div class=3D"gmail_extra"><span style=3D"color:=
rgb(34,34,34);font-family:arial,sans-serif;font-size:small;font-style:norma=
l;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;le=
tter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;wh=
ite-space:normal;word-spacing:0px;background-color:rgb(255,255,255);text-de=
coration-style:initial;text-decoration-color:initial;float:none;display:inl=
ine"><br></span></div><div class=3D"gmail_extra">Thu Apr=C2=A0 5 19:02:45 2=
018 VReadVolumeDiskHeader: Couldn't open header for volume 536881887 (e=
rrno 2).<br><div><br></div><div>but if that's such a problem then why c=
an the other AFS clients happily manipulate the volume?</div><div><br></div=
><div>It's puzzling.</div><span class=3D"HOEnZb"><font color=3D"#888888=
"><div><br></div><div>-Fred</div></font></span></div></div>
</blockquote></div><br></div>
--001a1149a0ea4c204305699ad5ec--