taroon-list@redhat.com
[Top] [All Lists]

RE: OCFS and Out Of Memory issue

Subject: RE: OCFS and Out Of Memory issue
From: "Frank, Ryan"
Date: Thu, 3 Feb 2005 14:04:11 -0500
Yes I did reboot.  The attached file I sent has the slabcache up to when
the OOM Kill happened I will paste the last few results from that...

--PASTE--
Thu Feb  3 12:26:26 EST 2005
slabinfo - version: 1.1 (SMP)
kmem_cache           112    112    244    7    7    1 : 1008  252
nfs_write_data         0      0    384    0    0    1 :  496  124
nfs_read_data          0      0    384    0    0    1 :  496  124
nfs_page               0      0    128    0    0    1 : 1008  252
ocfs_fileentry        96     96    512   12   12    1 :  496  124
ocfs_lockres         180    180    256   12   12    1 : 1008  252
ocfs_ofile           430    430    384   43   43    1 :  496  124
ocfs_oin             180    180    256   12   12    1 : 1008  252
ip_fib_hash           16    448     32    4    4    1 : 1008  252
ext3_xattr             0      0     44    0    0    1 : 1008  252
journal_head         280   1078     48   14   14    1 : 1008  252
revoke_table           6    750     12    3    3    1 : 1008  252
revoke_record          0      0     32    0    0    1 : 1008  252
clip_arp_cache         0      0    256    0    0    1 : 1008  252
ip_mrt_cache           0      0    128    0    0    1 : 1008  252
tcp_tw_bucket         12    150    128    5    5    1 : 1008  252
tcp_bind_bucket       49   1120     32   10   10    1 : 1008  252
tcp_open_request       0      0    128    0    0    1 : 1008  252
inet_peer_cache        4    174     64    3    3    1 : 1008  252
secpath_cache          0      0    128    0    0    1 : 1008  252
xfrm_dst_cache         0      0    256    0    0    1 : 1008  252
ip_dst_cache          60    225    256   15   15    1 : 1008  252
arp_cache              7     45    256    3    3    1 : 1008  252
flow_cache             0      0    128    0    0    1 : 1008  252
blkdev_requests    15360  15450    128  515  515    1 : 1008  252
kioctx                 0      0    128    0    0    1 : 1008  252
kiocb                  0      0    128    0    0    1 : 1008  252
dnotify_cache          0      0     20    0    0    1 : 1008  252
file_lock_cache       66    400     96   10   10    1 : 1008  252
async_poll_table       0      0    140    0    0    1 : 1008  252
fasync_cache           0      0     16    0    0    1 : 1008  252
uid_cache              9    560     32    5    5    1 : 1008  252
skbuff_head_cache    949   7015    168  305  305    1 : 1008  252
sock                2172   2255   1408  451  451    2 :  240   60
sigqueue             232    232    132    8    8    1 : 1008  252
kiobuf               270    720    128   24   24    1 : 1008  252
cdev_cache            18    406     64    7    7    1 : 1008  252
bdev_cache            20    348     64    6    6    1 : 1008  252
mnt_cache             26    348     64    6    6    1 : 1008  252
inode_cache        71214  77595    512 11085 11085    1 :  496  124
dentry_cache        4214   5940    128  198  198    1 : 1008  252
dquot                  0      0    128    0    0    1 : 1008  252
filp                5760   5970    128  199  199    1 : 1008  252
names_cache           23     23   4096   23   23    1 :  240   60
buffer_head       4935182 4936190    108 141009 141034    1 : 1008  252
mm_struct            285    390    384   39   39    1 :  496  124
vm_area_struct     10765  14000     68  250  250    1 : 1008  252
fs_cache             653   1102     64   19   19    1 : 1008  252
files_cache          296    378    512   54   54    1 :  496  124
signal_cache         452   1102     64   19   19    1 : 1008  252
sighand_cache        235    355   1408   71   71    2 :  240   60
pte_chain          60829  72300    128 2410 2410    1 : 1008  252
pae_pgd              647   1044     64   18   18    1 : 1008  252
size-131072(DMA)       0      0 131072    0    0   32 :    0    0
size-131072            0      0 131072    0    0   32 :    0    0
size-65536(DMA)        0      0  65536    0    0   16 :    0    0
size-65536             0      0  65536    0    0   16 :    0    0
size-32768(DMA)        0      0  32768    0    0    8 :    0    0
size-32768            13     13  32768   13   13    8 :    0    0
size-16384(DMA)        0      0  16384    0    0    4 :    0    0
size-16384            22     22  16384   22   22    4 :    0    0
size-8192(DMA)         0      0   8192    0    0    2 :    0    0
size-8192             44     44   8192   44   44    2 :    0    0
size-4096(DMA)         0      0   4096    0    0    1 :  240   60
size-4096           1079   1079   4096 1079 1079    1 :  240   60
size-2048(DMA)         0      0   2048    0    0    1 :  240   60
size-2048            262    262   2048  131  131    1 :  240   60
size-1024(DMA)         0      0   1024    0    0    1 :  496  124
size-1024            508    628   1024  157  157    1 :  496  124
size-512(DMA)          0      0    512    0    0    1 :  496  124
size-512             741    744    512   93   93    1 :  496  124
size-256(DMA)          0      0    256    0    0    1 : 1008  252
size-256            2220   2220    256  148  148    1 : 1008  252
size-128(DMA)          0      0    128    0    0    1 : 1008  252
size-128            3839   6510    128  217  217    1 : 1008  252
size-64(DMA)           0      0    128    0    0    1 : 1008  252
size-64              724    870    128   29   29    1 : 1008  252
size-32(DMA)           0      0     64    0    0    1 : 1008  252
size-32             3509   4640     64   80   80    1 : 1008  252
--END PASTE--

That was the last entry before the system stopped responding... If this
still does not have enough I can re-run the process again and capture...



--
-Ryan
Allvac Unix Services
Monroe, NC
(704) 282-1586

-----Original Message-----
From: taroon-list-bounces@xxxxxxxxxx
[mailto:taroon-list-bounces@xxxxxxxxxx] On Behalf Of Larry Woodman
Sent: Thursday, February 03, 2005 2:00 PM
To: Discussion of Red Hat Enterprise Linux 3 (Taroon)
Subject: Re: OCFS and Out Of Memory issue

Frank, Ryan wrote:

>Here is a paste of the current /proc/slabinfo
>
You must have rebooted your system, there is only about 7000 slabcache
pages now.  Please get the /proc/slabinfo output right at the OOM kill.

Larry

>
>--BEGIN PASTE--
>slabinfo - version: 1.1 (SMP)
>kmem_cache           144    144    244    9    9    1 : 1008  252
>nfs_write_data        30     30    384    3    3    1 :  496  124
>nfs_read_data         10     10    384    1    1    1 :  496  124
>nfs_page              60     60    128    2    2    1 : 1008  252
>ocfs_fileentry        72     72    512    9    9    1 :  496  124
>ocfs_lockres          30     30    256    2    2    1 : 1008  252
>ocfs_ofile            80     80    384    8    8    1 :  496  124
>ocfs_oin              45     45    256    3    3    1 : 1008  252
>ip_fib_hash          560    560     32    5    5    1 : 1008  252
>ext3_xattr             0      0     44    0    0    1 : 1008  252
>journal_head        1309   1309     48   17   17    1 : 1008  252
>revoke_table         750    750     12    3    3    1 : 1008  252
>revoke_record        448    448     32    4    4    1 : 1008  252
>clip_arp_cache         0      0    256    0    0    1 : 1008  252
>ip_mrt_cache           0      0    128    0    0    1 : 1008  252
>tcp_tw_bucket        120    120    128    4    4    1 : 1008  252
>tcp_bind_bucket      896    896     32    8    8    1 : 1008  252
>tcp_open_request      60     60    128    2    2    1 : 1008  252
>inet_peer_cache      116    116     64    2    2    1 : 1008  252
>secpath_cache          0      0    128    0    0    1 : 1008  252
>xfrm_dst_cache         0      0    256    0    0    1 : 1008  252
>ip_dst_cache         360    360    256   24   24    1 : 1008  252
>arp_cache             45     45    256    3    3    1 : 1008  252
>flow_cache             0      0    128    0    0    1 : 1008  252
>blkdev_requests    15420  15420    128  514  514    1 : 1008  252
>kioctx                 0      0    128    0    0    1 : 1008  252
>kiocb                  0      0    128    0    0    1 : 1008  252
>dnotify_cache          0      0     20    0    0    1 : 1008  252
>file_lock_cache      320    320     96    8    8    1 : 1008  252
>async_poll_table       0      0    140    0    0    1 : 1008  252
>fasync_cache           0      0     16    0    0    1 : 1008  252
>uid_cache            784    784     32    7    7    1 : 1008  252
>skbuff_head_cache   6004   6256    168  272  272    1 : 1008  252
>sock                 375    375   1408   75   75    2 :  240   60
>sigqueue             290    290    132   10   10    1 : 1008  252
>kiobuf               240    240    128    8    8    1 : 1008  252
>cdev_cache          2900   2900     64   50   50    1 : 1008  252
>bdev_cache           464    464     64    8    8    1 : 1008  252
>mnt_cache            290    290     64    5    5    1 : 1008  252
>inode_cache        23289  23289    512 3327 3327    1 :  496  124
>dentry_cache       25350  25350    128  845  845    1 : 1008  252
>dquot                  0      0    128    0    0    1 : 1008  252
>filp                 870    870    128   29   29    1 : 1008  252
>names_cache           28     28   4096   28   28    1 :  240   60
>buffer_head        11200  11200    108  320  320    1 : 1008  252
>mm_struct            290    290    384   29   29    1 :  496  124
>vm_area_struct      5264   5264     68   94   94    1 : 1008  252
>fs_cache             522    522     64    9    9    1 : 1008  252
>files_cache          315    315    512   45   45    1 :  496  124
>signal_cache         696    696     64   12   12    1 : 1008  252
>sighand_cache        455    455   1408   91   91    2 :  240   60
>pte_chain           6366   6870    128  229  229    1 : 1008  252
>pae_pgd              580    580     64   10   10    1 : 1008  252
>size-131072(DMA)       0      0 131072    0    0   32 :    0    0
>size-131072            0      0 131072    0    0   32 :    0    0
>size-65536(DMA)        0      0  65536    0    0   16 :    0    0
>size-65536             0      0  65536    0    0   16 :    0    0
>size-32768(DMA)        0      0  32768    0    0    8 :    0    0
>size-32768            11     12  32768   11   12    8 :    0    0
>size-16384(DMA)        0      0  16384    0    0    4 :    0    0
>size-16384            22     23  16384   22   23    4 :    0    0
>size-8192(DMA)         0      0   8192    0    0    2 :    0    0
>size-8192             21     23   8192   21   23    2 :    0    0
>size-4096(DMA)         0      0   4096    0    0    1 :  240   60
>size-4096           2297   2477   4096 2297 2477    1 :  240   60
>size-2048(DMA)         0      0   2048    0    0    1 :  240   60
>size-2048            720    780   2048  360  390    1 :  240   60
>size-1024(DMA)         0      0   1024    0    0    1 :  496  124
>size-1024           1056   1056   1024  264  264    1 :  496  124
>size-512(DMA)          0      0    512    0    0    1 :  496  124
>size-512            1288   1288    512  161  161    1 :  496  124
>size-256(DMA)          0      0    256    0    0    1 : 1008  252
>size-256            1785   3045    256  152  203    1 : 1008  252
>size-128(DMA)          0      0    128    0    0    1 : 1008  252
>size-128            4590   4590    128  153  153    1 : 1008  252
>size-64(DMA)           0      0    128    0    0    1 : 1008  252
>size-64              570    570    128   19   19    1 : 1008  252
>size-32(DMA)           0      0     64    0    0    1 : 1008  252
>size-32             2802   3306     64   57   57    1 : 1008  252
>--END PASTE--
>
>Also attached is the slabinfo results while running the `dd` (requested

>by RedHat)
>
>--
>-Ryan (RHCE, SCSA)
>Allvac Unix Services
>Monroe, NC
>(704) 282-1586
>
>-----Original Message-----
>From: taroon-list-bounces@xxxxxxxxxx
>[mailto:taroon-list-bounces@xxxxxxxxxx] On Behalf Of Larry Woodman
>Sent: Thursday, February 03, 2005 1:35 PM
>To: Discussion of Red Hat Enterprise Linux 3 (Taroon)
>Subject: Re: OCFS and Out Of Memory issue
>
>Frank, Ryan wrote:
>
>  
>
>>Hello all!
>>
>>I am using AS 3.0 update 3 (Kernel 2.4.21-20.Elsmp) on a pair of 
>>Compaq DL-580's with 12gb of RAM. We are running Oracle 10g release 3,
>>    
>>
>
>  
>
>>in a RAC configuration. I am using SAN storage (Hitachi 9585) via 
>>Emulex LP9002 HBA's. Cluster file system is provided by OCFS 1.0.13.
>>
>>The problem is like this... When I try to create a large file on an
>>    
>>
>OCFS
>  
>
>>volume I get the following error from the system when the cached 
>>memory approaches 4GB.
>>
>>    
>>
>
>The problem here is the slab cache has consumed all of the Normal 
>memory
>zone:
>
> >>>Feb 3 11:53:54 alvmnrlnx004 kernel: aa:0 ac:340 id:2 il:74 ic:0
>fr:617  >>>Feb 3 11:53:54 alvmnrlnx004 kernel: 161776 pages of 
>slabcache
>
>Please send along a "cat /proc/slabinfo" output.
>
>Larry Woodman
>
>
>--
>Taroon-list mailing list
>Taroon-list@xxxxxxxxxx
>http://www.redhat.com/mailman/listinfo/taroon-list
>
>  
>
>-----------------------------------------------------------------------
>-
>
>--
>Taroon-list mailing list
>Taroon-list@xxxxxxxxxx
>http://www.redhat.com/mailman/listinfo/taroon-list
>


--
Taroon-list mailing list
Taroon-list@xxxxxxxxxx
http://www.redhat.com/mailman/listinfo/taroon-list


--
Taroon-list mailing list
Taroon-list@xxxxxxxxxx
http://www.redhat.com/mailman/listinfo/taroon-list

<Prev in Thread] Current Thread [Next in Thread>