Dear all,
i forward the following mail to ltsp and beowulf as I use these concepts and
Mosix group seems to be in a very depressed mood.
The problem I encounter occurs before Mosix even starts. There is some new
kind of stuff with kernel 2.4.xx that does not accept some kinds of fragments.
It is more an NFS boot problem.
One (quick and dirty) solution could be to allow the kernel to load even with
an mtu less than 1500, which I could not do.
Thanks in advance
Jacques
--
Dr. Jacques B. Siboni mailto:[EMAIL PROTECTED]
8 pass. Charles Albert, F75018 Paris, France
Tel. & Fax: 33 (0) 1 42 28 76 78
Home Page: http://www.lutecium.org/jacsib/
--- Begin Message ---
Hi all,
Since I installed Mosix 1.5.2 with linux 2.4.13 I encounter a problem. NFS is
not happy and tcpdump says
icmp: ip reassembly time exceeded
The mtu size is 1500 for eth1
If after kernel boot I change mtu to 600, NFS is happy again and NFS mounts
resume ok.
BUT If I put mtu down to 600 before dhcp kernel boot. I get complains on the
node machine saying:
ALERT: got a fragmented packet - reconfigure your server
and kernel never loads
I can't switch mtu size during boot process!
The funny thing is that boot process is ok with mtu 1500 and an old 2.2.19
version on the slave node although the server is in 2.4.13.
There must be some options to change in the kernel config that could help.
What do you suggest
Thanks in advance
Happy new year
Jacques
--
Dr. Jacques B. Siboni mailto:[EMAIL PROTECTED]
8 pass. Charles Albert, F75018 Paris, France
Tel. & Fax: 33 (0) 1 42 28 76 78
Home Page: http://www.lutecium.org/jacsib/
--- End Message ---
