[OpenAFS-devel] Final Mac OS Steps
Derrick Brashear
shadow@gmail.com
Tue, 31 Mar 2009 13:03:50 -0400
--000e0cd5d022d80fa004666d2e41
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
On Tue, Mar 31, 2009 at 12:36 PM, Barry Fawthrop <barry@isscp.com> wrote:
> To All,
>
> Not Sure how or to who: But setting the LWP_SP to 9 for the iPhone
> simulator
> definitely works, the AFS_DARWIN_ENV sets it to 16, in process.c
> perhaps we can create a AFS_DARWIN_ARM_ENV for the iPhone ????
>
>
i'd suggest AFS_ARM_ENV and AFS_DARWIN_ENV combined, as that's already how
PPC and X86 are done.
>
> In trying to trace the recvmsg problem (for the reply of NameToID)
> Tracing with wireshark I see two issues
>
> (1) Every packets sent from my client to the server has a Checksum
> incorrect
> maybe caused by "UDP checksum offload"? message.
> Not sure what is causing this
>
that won't matter.
>
>
>
> (2) The kerberos AS-REQ/AS-REP and TGS-REQ/TGS-REP work fine
> But the client sends to the server the
> AFS (RX) PROT Request: name-to-id (504)
> The server sends back multiple copies (about 4) responses
> But obviously the client has broken out in the error / hung up
>
it sends multiple replies because it's not getting an ack.
if recvmsg on the iPhone doesn't believe it's receiving, it will never ack,
so...
--
Derrick
--000e0cd5d022d80fa004666d2e41
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
<br><br><div class=3D"gmail_quote">On Tue, Mar 31, 2009 at 12:36 PM, Barry =
Fawthrop <span dir=3D"ltr"><<a href=3D"mailto:barry@isscp.com">barry@iss=
cp.com</a>></span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"=
border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; paddi=
ng-left: 1ex;">
To All,<br>
<br>
Not Sure how or to who: =A0 But setting the LWP_SP to 9 for the iPhone simu=
lator<br>
definitely works, the AFS_DARWIN_ENV =A0sets it to 16, in process.c<br>
perhaps we can create a AFS_DARWIN_ARM_ENV for the iPhone ????<br>
<br>
</blockquote><div><br>i'd suggest AFS_ARM_ENV and AFS_DARWIN_ENV combin=
ed, as that's already how PPC and X86 are done.<br>=A0<br></div><blockq=
uote class=3D"gmail_quote" style=3D"border-left: 1px solid rgb(204, 204, 20=
4); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<br>
In trying to trace the recvmsg problem (for the reply of NameToID)<br>
Tracing with wireshark I see two issues<br>
<br>
(1) Every packets sent from my client to the server has a Checksum incorrec=
t<br>
maybe caused by "UDP checksum offload"? =A0message.<br>
Not sure what is causing this<br>
</blockquote><div><br>that won't matter.<br>=A0<br></div><blockquote cl=
ass=3D"gmail_quote" style=3D"border-left: 1px solid rgb(204, 204, 204); mar=
gin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><br>
<br>
<br>
(2) The kerberos AS-REQ/AS-REP =A0and TGS-REQ/TGS-REP =A0work fine<br>
But the client sends to the server the<br>
AFS (RX) PROT Request: name-to-id (504)<br>
The server sends back multiple copies (about 4) responses<br>
But obviously the client has broken out in the error / hung up<br>
</blockquote><div><br>it sends multiple replies because it's not gettin=
g an ack.=A0 <br>if recvmsg on the iPhone doesn't believe it's rece=
iving, it will never ack, so...<br></div></div><br>-- <br>Derrick<br>
--000e0cd5d022d80fa004666d2e41--