Re: Lockup 2.1.6* during network activity

Tony Lill (ajlill@ajlc.waterloo.on.ca)
Thu, 06 Nov 1997 17:52:44 EST


I've been getting that on all the .50 and .60 series kernels, and
occaisonally on .36. Are you getting a lot of

fork: Out of memory
IP: queue_glue: no memory for gluing queue c0afe2e0
RPC: buffer allocation failed for task c0a83770

and assorted other messages of the ilk. Sometimes it recovers,
sometime not. Sometimes everything dies but init. It's kind of neat to
come back to the system and find nothing running but init and the gettys.

Here's a meminfo from the last one. Swap's down to 8M, but I've had
it happen while I still have 30-40M of swap left, I think. In case I'm
out to lunch and just running out of swap, I've tuned down http and
doubled my swap to see if it goes away.

Mem-info:
Free pages: 300kB
( 29*4kB 7*8kB 8*16kB 0*32kB 0*64kB 0*128kB = 300kB)
Swap cache: add 84624/84624, delete 46217129/56707, find 188619/27628
Free swap: 8176kB
4080 pages of RAM
499 free pages
412 reserved pages
3591 pages shared
Buffer memory: 56kB
Buffer heads: 218
Buffer blocks: 56
CLEAN: 20 buffers, 5 used (last=19), 5 locked, 0 protected, 0 dirty
LOCKED: 16 buffers, 16 used (last=16), 0 locked, 0 protected, 0 dirty
DIRTY: 10 buffers, 3 used (last=10), 0 locked, 0 protected, 10 dirty
Networking buffers in use : 1946
Total network buffer allocations : 4443120
Total failed network buffer allocs : 798133
IP fragment buffer size : 0

--
Tony Lill,                         Tony.Lill@AJLC.Waterloo.ON.CA
President, A. J. Lill Consultants        fax/data (519) 650 3571
539 Grand Valley Dr., Cambridge, Ont. N3H 2S2     (519) 241 2461
--------------- http://www.ajlc.waterloo.on.ca/ ----------------
"Welcome to All Things UNIX, where if it's not UNIX, it's CRAP!"