Am Samstag, den 09.08.2008, 13:52 +0100 schrieb David Howells: > Thomas Meyer <thomas@xxxxxxxx> wrote: > > > > > Beginning with next-20080808 after letting the system run for 5 minutes > > > > or so, I get an error from the fork call, e.g.: > > > > > > > > "bash: fork: Resource temporarily unavailable" > > I modified your config slightly so that it'd run on my testbox (x86_64 > unfortunately), but that survived LTP and seemed to work okay. I'll have to > snag an i386 installation from somewhere, unless James can do me a favour and > test it on his, if he has one. > > Can you try this please: > > cat /proc/slabinfo | cut -d: -f1 | sort -k 2 -n > > Just to check to see if there's a memory leak. I'm too lazy to reboot now. But i can give you that from vmcore dump: crash> kmem -i PAGES TOTAL PERCENTAGE TOTAL MEM 1524251 5.8 GB ---- FREE 1343484 5.1 GB 88% of TOTAL MEM USED 180767 706.1 MB 11% of TOTAL MEM SHARED 19076 74.5 MB 1% of TOTAL MEM BUFFERS 3800 14.8 MB 0% of TOTAL MEM CACHED 108483 423.8 MB 7% of TOTAL MEM SLAB 6239 24.4 MB 0% of TOTAL MEM TOTAL HIGH 1343440 5.1 GB 88% of TOTAL MEM FREE HIGH 0 0 0% of TOTAL HIGH TOTAL LOW 180811 706.3 MB 11% of TOTAL MEM FREE LOW 1343484 5.1 GB 743% of TOTAL LOW TOTAL SWAP 1596751 6.1 GB ---- SWAP USED 0 0 0% of TOTAL SWAP SWAP FREE 1596751 6.1 GB 100% of TOTAL SWAP Is that what you wanted to know? Doesn't look like a memory leak for me. kmem -s doesn't seem to work: crash> kmem -s kmem: invalid structure member offset: kmem_cache_objects FILE: memory.c LINE: 13500 FUNCTION: dump_kmem_cache_slub() [/usr/bin/crash] error trace: 8085195 => 809f8cf => 80c0d82 => 813b9a0 CACHE NAME OBJSIZE ALLOCATED TOTAL SLABS SSIZE 813b9a0: OFFSET_verify+118 80c0d82: dump_kmem_cache_slub+1033 809f8cf: cmd_kmem+3077 8085195: exec_command+265 kmem: invalid structure member offset: kmem_cache_objects FILE: memory.c LINE: 13500 FUNCTION: dump_kmem_cache_slub() -- To unsubscribe from this list: send the line "unsubscribe linux-next" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html