http://oss.sgi.com/bugzilla/show_bug.cgi?id=266
Summary: Redhat8: NFS transfers lockup server
Product: Linux XFS
Version: 1.2.x
Platform: All
OS/Version: Linux
Status: NEW
Severity: critical
Priority: High
Component: XFS kernel code
AssignedTo: xfs-master@xxxxxxxxxxx
ReportedBy: darstout@xxxxxxxxxxx
Hello,
I have a dual processor (Xeon 2.2) Gigabyte motherboard w/1MB mem running a
3ware Escalade 7500 driving 6 Maxtor 200GB hard drives. The O/S is RH8 loaded
from the ISO image for XFS 1.2. The 6 drives are configured as a 1TB filesystem.
I filesystem was built with the command "mkfs.xfs /dev/sda1". The NFS partner
machine that I am "talking" to is a Sun 880 (2 procs/4GB mem). Everything is ok
when I am transferring data from the Sun to the RH machine (over 60GB/hr)
without a single error listed in /var/log/messages BUT when I try to "pull" data
"from" the SUN (about 100-400GB) I will get a hang on the RH machine about half
way through. There are no error messages in /var/log/messages on RH and no
error messages on the Sun side. The RH machine cannot be contacted by the
console or by the network. The data rates are very high (over 80GB/hr) when
pulling the data. Could this be a buffer overflow somewhere? Do I have to
adjust any parameters anywhere?
Please help.
Thank You.
Darius
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
|