I have several Dell WS410 dual PII-400 workstations being used as
a part of a 25 node cluster. With the new 2.2.0 kernels, I get the
following errors (listed at the end of the mail) during bootup and
consequently my overall network performance goes down by about a factor of
2 (for my MPI codes atleast)! I am also unable to login to the node via
the network for approximately 3-4 minutes after the bootup has finished.
All the user hard disk space is NFS mounted on every node. Earlier my
3c905B was not running properly with the 3c59x.c (v0.99E) supplied with
the kernel, but on obtaining v0.99H from:
http://cesdis.gsfc.nasa.gov/linux/drivers/3c59x.c
the problem disappeared and my card runs at 100Mbps full-duplex mode)!
The same kernel however runs fine on the server which is also the exact
same hardware except for more hard disk space! I am using RedHat 5.1. The
options compiled into the kernel can be known from the file:
http://bart.aero.psu.edu/linux/temp/config
Also, I use egcs-2.90.27 980315 with "-mpentiumpro -O6-fno-inline-functions").
Any clues as to what is happening ?? Thanx in advance and do let me know
if any additional info is needed. The same problem continues with
2.2.0-pre2 which I just tested with the additional problem that my
3c905B only runs in half-duplex mode even with v0.99H drivers.
The error:
----------
Dec 31 10:51:29 node23 kernel: RPC: task of released request still queued!
Dec 31 10:51:29 node23 kernel: RPC: (task is on xprt_pending)
Dec 31 10:51:29 node23 kernel: lockd_up: no pid, 2 users??
Dec 31 10:51:29 node23 kernel: portmap: RPC call returned error 111
Dec 31 10:51:29 node23 kernel: RPC: task of released request still queued!
Dec 31 10:51:29 node23 kernel: RPC: (task is on xprt_pending)
Dec 31 10:51:29 node23 kernel: portmap: RPC call returned error 111
Dec 31 10:51:29 node23 kernel: RPC: task of released request still queued!
Dec 31 10:51:29 node23 kernel: RPC: (task is on xprt_pending)
Dec 31 10:51:29 node23 kernel: portmap: RPC call returned error 111
Dec 31 10:51:29 node23 kernel: RPC: task of released request still queued!
Dec 31 10:51:29 node23 kernel: RPC: (task is on xprt_pending)
Dec 31 10:51:29 node23 kernel: lockd_up: no pid, 3 users??
Dec 31 10:51:29 node23 kernel: portmap: RPC call returned error 111
Dec 31 10:51:29 node23 kernel: RPC: task of released request still queued!
Dec 31 10:51:29 node23 kernel: RPC: (task is on xprt_pending)
Dec 31 10:51:29 node23 kernel: portmap: RPC call returned error 111
Dec 31 10:51:29 node23 kernel: RPC: task of released request still queued!
Dec 31 10:51:29 node23 kernel: RPC: (task is on xprt_pending)
Dec 31 10:51:29 node23 kernel: portmap: RPC call returned error 111
Dec 31 10:51:29 node23 kernel: RPC: task of released request still queued!
Dec 31 10:51:29 node23 kernel: RPC: (task is on xprt_pending)
Dec 31 10:51:29 node23 kernel: lockd_up: no pid, 4 users??
-anirudh
[http://bart.aero.psu.edu]
Our cluster COCOA: http://cocoa.ihpca.psu.edu
---
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/