The problem appears when I write big files relatively fast over
nfs. My client is Dual-Athlon 1900+, and the servers are Dual P3
500MHz (with 2.4.18) and Pentium 2 200MHz (with 2.4.18 debian bf-2.4
default kernel).
Both servers use kernel-nfsd. However, as things work perfectly with a
2.4.19-client and between the servers, it would hardly seem to be a
server-matter.
There is a gigabit-link between the Athlon and P3.
Reproducing: dd if=/dev/zero of=new-file-on-nfs-file-system -> dies on
I/O error.
If I dd if=/dev/zero bs=1024 count=1000000 of=existing-one-GB-file,
things fail similarly. (Last attempt failed at record 3073.)
It might be just luck, but it would seem to me that the Dual-P3 server
behind the gigabit-link fails faster than the slow P2-server (which is
infact network-topographically behind the P3).
Relevant lines from .config:
CONFIG_NFS_FS=m
CONFIG_NFS_V3=y
# CONFIG_ROOT_NFS is not set
CONFIG_NFSD=m
CONFIG_NFSD_V3=y
# CONFIG_NFSD_TCP is not set
CONFIG_SUNRPC=m
CONFIG_LOCKD=m
CONFIG_LOCKD_V4=y
I'm willing to try patches :).
-- _____________________________________________________________________ / __// /__ ____ __ http://www.modeemi.fi/~flux/\ \ / /_ / // // /\ \/ / \ / /_/ /_/ \___/ /_/\_\@modeemi.fi \/ - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/