Anyway, problem seem to have gone away..
Machine is really slow while trying to do IO on
all 4000 filesystems. It is slowly creating processes.
System is 100% busy. (so far, it created 461 processes
out of 4000). But we are not running out of lowmem and
inode caches seems to be reasonable..
So, it must be some leak in ext3..
procs memory swap io system cpu
r b w swpd free buff cache si so bi bo in cs us sy id
407 31 8 0 3334068 80996 150144 0 0 148795 12 1018 18807 0 100 0
413 25 9 0 3334004 80996 150144 0 0 149867 20 1014 18910 0 100 0
411 27 8 0 3334068 80996 150144 0 0 154293 26 1017 19447 0 100 0
405 33 8 0 3333940 80996 150144 0 0 156830 14 1013 19747 0 100 0
400 38 8 0 3334004 80996 150144 0 0 154778 9 1013 19525 0 100 0
425 18 8 0 3333180 81040 150100 0 0 155303 8 1016 19615 0 100 0
409 34 8 0 3333116 81256 150144 0 0 151814 13 1015 19182 0 100 0
416 30 9 0 3332564 81396 150264 0 0 134068 10 1015 17024 0 100 0
[root@elm3b78 root]# cat /proc/meminfo
MemTotal: 3883248 kB
MemFree: 3330836 kB
Buffers: 82328 kB
Cached: 150112 kB
SwapCached: 0 kB
Active: 129736 kB
Inactive: 141412 kB
HighTotal: 3014616 kB
HighFree: 2797632 kB
LowTotal: 868632 kB
LowFree: 533204 kB <<<<
SwapTotal: 2040244 kB
SwapFree: 2040244 kB
Dirty: 1372 kB
Writeback: 0 kB
Mapped: 45224 kB
Slab: 231516 kB
Committed_AS: 47420 kB
PageTables: 5884 kB
ReverseMaps: 9475
[root@elm3b78 root]# grep inode /proc/slabinfo
rpc_inode_cache 0 0 404 0 0 1 : 32 16 : 0 0 0 0 0 0 153 : 0 0 0 0
udf_inode_cache 0 0 420 0 0 1 : 32 16 : 0 0 0 0 0 0 153 : 0 0 0 0
nfs_inode_cache 0 0 636 0 0 1 : 32 16 : 0 0 0 0 0 0 150 : 0 0 0 0
isofs_inode_cache 0 0 380 0 0 1 : 32 16 : 0 0 0 0 0 0 154 : 0 0 0 0
ext2_inode_cache 11425 11432 480 1429 1429 1 : 32 16 : 11432 11740 1429 0 0 0 152 : 9948 1487 10 0\
ext3_inode_cache 4720 4728 496 591 591 1 : 32 16 : 4728 4850 591 0 0 0 152 : 8147 610 4037 0
(due to mount points /mnt/mnt1 /mnt/mnt2 .. /mnt/mnt4000)
shmem_inode_cache 2 8 464 1 1 1 : 32 16 : 8 15 1 0 0 0 152 : 0 2 0 0
sock_inode_cache 38 45 408 5 5 1 : 32 16 : 45 179 5 0 0 0 153 : 373 22 357 0
proc_inode_cache 2810 2810 380 281 281 1 : 32 16 : 2810 5428 304 23 0 0 154 : 3759 553 1519 0
inode_cache 148260 148270 364 14827 14827 1 : 32 16 : 148270 148474 14829 2 0 0 154 : 141839 15202 8781 0
(due to /sysfs)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/