My standard stress test here is to run make -j on the tk8.0 sources mounted
via nfs. That typically runs well into swap on my machine and drives the load
average up to around 100.
It completes without crashing anything these days but ...
With the heavy swapping going on I'm obviously having problems getting memory
for network buffers somewhere.
IP: queue_glue: no memory for gluing queue c2fa70c0
RPC: task timed out
IP: queue_glue: no memory for gluing queue c2fa70c0
RPC: task timed out
IP: queue_glue: no memory for gluing queue c2fa70c0
RPC: task timed out
IP: queue_glue: no memory for gluing queue c2fa70c0
RPC: task timed out
IP: queue_glue: no memory for gluing queue c2fa70c0
RPC: task timed out
IP: queue_glue: no memory for gluing queue c2fa70c0
RPC: task timed out
IP: queue_glue: no memory for gluing queue c2fa70c0
RPC: task timed out
nfs: server ongaonga OK
Is there any simple way to tweak the system to reduce this problem ?
Thanks
Peter
----------------------------------
E-Mail: P.Waltenberg@irl.cri.nz
Date: 23-Feb-98
Time: 09:00:04
----------------------------------
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu