So a bad network card (or maybe pci-X slot) is causing corruption that
still has a correct TCP checksum?
Can you run tcpdump on both server and client and save a trace?
You could also try turning off any TCP or checksum offload in the
network card.
Emmanuel Florac wrote:
Le Fri, 30 May 2008 13:19:36 +0200
Emmanuel Florac <[EMAIL PROTECTED]> écrivait:
Can it be the origin of the culprit? Some buffering error on the
machine with the smallest memory size ?
Actually it's a network problem. I'm using PCI-X Intel Pro-1000 MT dual
ports cards to interconnect the cluster nodes. I've switched from using
a port (eth2) to the other one (eth3) and it got ... much worse!
cluster2 read complete garbage now, while cluster1 still does OK.
I'm going to try replacing the network card in cluster2.
_______________________________________________
Pvfs2-users mailing list
[email protected]
http://www.beowulf-underground.org/mailman/listinfo/pvfs2-users
