data transfer "freeze up" using pptpd/ppp (VPN)

[Hawkeye]

Packages and versions:
2.6.18-1.2869.fc6
ppp-2.4.4-1.fc6
pptpd-1.3.3-1.fc6


options.pptpd:
require-mschap-v2
require-mppe-128
proxyarp
debug
lock

nobsdcomp
novj
novjccomp

nologfd

auth
nodefaultroute
plugin winbind.so
ntlm_auth-helper "/usr/bin/ntlm_auth --helper-protocol=ntlm-server-1
--require-membership-of=LANDOFMILKHONEY+RemoteAccess"

I have no issues connecting nor do I have a problem on the initial
connection of resources over the VPN such as SSH/FTP/HTTP/RDP/SMB but
the moment I start generating a pile of traffic, aka large file
transfer, my vpn connection seizes up and all I see in the logs are
piles of ...

Jan 8 15:41:37 edge pppd[10836]: rcvd [proto=0xeae8] 8b bd 9d 01 73 b5
4e 0d a1 ac 8e e8 15 84 79 e2 a5 24 49 43 af 4b eb d1 51 a4 ee 5b ae b2
8
d d1 ...
Jan 8 15:41:37 edge pppd[10836]: Unsupported protocol 0xeae8 received
Jan 8 15:41:37 edge pppd[10836]: Unsupported protocol 0xeae8 received
Jan 8 15:41:37 edge pppd[10836]: sent [LCP ProtRej id=0x2 ea e8 8b bd
9d 01 73 b5 4e 0d a1 ac 8e e8 15 84 79 e2 a5 24 49 43 af 4b eb d1 51 a4
ee
5b ae b2 ...]

I have to hang up my vpn session and re-establish it to get
connectivity again. As long as I do not generate a pile of traffic, it
is stable, does anyone know why?

[Unruh]

Post by Hawkeye
2.6.18-1.2869.fc6
ppp-2.4.4-1.fc6
pptpd-1.3.3-1.fc6
require-mschap-v2
require-mppe-128
proxyarp

Those all look like you are the ISP fro a remote system. Is that right?

Post by Hawkeye
debug
lock
nobsdcomp
novj
novjccomp
nologfd
auth

As does this

Post by Hawkeye
nodefaultroute
plugin winbind.so
ntlm_auth-helper "/usr/bin/ntlm_auth --helper-protocol=ntlm-server-1
--require-membership-of=LANDOFMILKHONEY+RemoteAccess"
I have no issues connecting nor do I have a problem on the initial
connection of resources over the VPN such as SSH/FTP/HTTP/RDP/SMB but
the moment I start generating a pile of traffic, aka large file
transfer, my vpn connection seizes up and all I see in the logs are
piles of ...

You need to show us the negotiation with the other end and the beginning of
the problem. The system should not be restarting ppp negotiations.
I suspect that your do not escape the right things. Eg
asyncmap 0xa0000
might help as an option.

Post by Hawkeye
Jan 8 15:41:37 edge pppd[10836]: rcvd [proto=0xeae8] 8b bd 9d 01 73 b5
4e 0d a1 ac 8e e8 15 84 79 e2 a5 24 49 43 af 4b eb d1 51 a4 ee 5b ae b2

It is interpreting the stream as ppp commands. It should not be.

Post by Hawkeye
8
d d1 ...
Jan 8 15:41:37 edge pppd[10836]: Unsupported protocol 0xeae8 received
Jan 8 15:41:37 edge pppd[10836]: Unsupported protocol 0xeae8 received
Jan 8 15:41:37 edge pppd[10836]: sent [LCP ProtRej id=0x2 ea e8 8b bd
9d 01 73 b5 4e 0d a1 ac 8e e8 15 84 79 e2 a5 24 49 43 af 4b eb d1 51 a4
ee
5b ae b2 ...]
I have to hang up my vpn session and re-establish it to get
connectivity again. As long as I do not generate a pile of traffic, it
is stable, does anyone know why?

[apearse]

Post by Unruh
Those all look like you are the ISP fro a remote system. Is that right?

Yes this is the server side.

Post by Unruh
You need to show us the negotiation with the other end and the beginning of
the problem. The system should not be restarting ppp negotiations.
I suspect that your do not escape the right things. Eg
asyncmap 0xa0000
might help as an option.

I just tried the asyncmap 0xa0000 option and it made no difference. The
moment I attempt any large sustained transfer, I am swamped with those
errors. If I can keep the traffic minimal, like pings, or even RDP, I
can keep the connection up for hours.

[Unruh]

Post by apearse

Post by Unruh
Those all look like you are the ISP fro a remote system. Is that right?

Yes this is the server side.

Post by Unruh
You need to show us the negotiation with the other end and the beginning of
the problem. The system should not be restarting ppp negotiations.
I suspect that your do not escape the right things. Eg
asyncmap 0xa0000
might help as an option.

I just tried the asyncmap 0xa0000 option and it made no difference. The
moment I attempt any large sustained transfer, I am swamped with those
errors. If I can keep the traffic minimal, like pings, or even RDP, I
can keep the connection up for hours.

As I said, show us the negotiations at the beginning of the session up to
the point where the pppd session begins to show those weird entries. The
ppp negotiations should not be open. Something is opening them again, or
they never closed.

[apearse]

I have had to drop this approach and installed windows 2003 and had it
up and running within 1.5 hours without a problem since. Thanks for the
feedback and suggestions though. I appreciated the support.