xfs_db: out of memory (2.9.8)

xfs_db: out of memory (2.9.8)

[Justin Piszcz]

For the root partition I get "out of memory" when I try to check for 
fragmentation.

# xfs_db -c frag -f /dev/md2
xfs_db: out of memory

# xfs_db -V
xfs_db version 2.9.8

# mount | grep "/dev/md2"
/dev/md2 on / type xfs (rw,noatime)

# xfs_info /dev/md2
meta-data=/dev/root              isize=256    agcount=16, agsize=2024942 blks
          =                       sectsz=512   attr=0
data     =                       bsize=4096   blocks=32399072, imaxpct=25
          =                       sunit=0      swidth=0 blks
naming   =version 2              bsize=4096 
log      =internal               bsize=4096   blocks=15819, version=1
          =                       sectsz=512   sunit=0 blks, lazy-count=0
realtime =none                   extsz=4096   blocks=0, rtextents=0

# xfs_db -c frag -f /dev/md3
actual 228842, ideal 223767, fragmentation factor 2.22%

Any idea?

Justin.

Re: xfs_db: out of memory (2.9.8)

[Eric Sandeen]

Justin Piszcz wrote:
> For the root partition I get "out of memory" when I try to check for 
> fragmentation.
> 
> # xfs_db -c frag -f /dev/md2
> xfs_db: out of memory

"-f" means "stored in a file"

> # mount | grep "/dev/md2"
> /dev/md2 on / type xfs (rw,noatime)

It looks like you're cheating your way around:

xfs_db: /dev/md2 contains a mounted filesystem

fatal error -- couldn't initialize XFS library

I'm not sure if that matters in this respect, though.

You could use gdb to find out which allocation is failing.

> # xfs_db -V
> xfs_db version 2.9.8
> 

> Any idea?

Don't worry so much about that fragmentation stat? :)

-Eric

> Justin.
> 
> 

Re: xfs_db: out of memory (2.9.8)

[Justin Piszcz]

On Sat, 25 Oct 2008, Eric Sandeen wrote:

> Justin Piszcz wrote:
>> For the root partition I get "out of memory" when I try to check for
>> fragmentation.
>>
>> # xfs_db -c frag -f /dev/md2
>> xfs_db: out of memory
>
> "-f" means "stored in a file"
>
>> # mount | grep "/dev/md2"
>> /dev/md2 on / type xfs (rw,noatime)
>
> It looks like you're cheating your way around:
>
> xfs_db: /dev/md2 contains a mounted filesystem
>
> fatal error -- couldn't initialize XFS library
>
> I'm not sure if that matters in this respect, though.
>
> You could use gdb to find out which allocation is failing.
>
>> # xfs_db -V
>> xfs_db version 2.9.8
>>
>
>> Any idea?
>
> Don't worry so much about that fragmentation stat? :)
>
> -Eric
>
>> Justin.
>>
>>
>

.. but it used to work :(

Loaded symbols for /lib/libgcc_s.so.1
Error while mapping shared library sections:
ease: No such file or directory.
(no debugging symbols found)
Core was generated by `xfs_db -c frag -r /dev/md2'.
Program terminated with signal 6, Aborted.
[New process 17097]
#0  0x00007f913f9b0ef5 in raise () from /lib/libc.so.6
(gdb) bt
#0  0x00007f913f9b0ef5 in raise () from /lib/libc.so.6
#1  0x00007f913f9b2413 in abort () from /lib/libc.so.6
#2  0x00007f913f9ed3e8 in ?? () from /lib/libc.so.6
#3  0x00007f913f9f2968 in ?? () from /lib/libc.so.6
#4  0x00007f913f9f4a76 in free () from /lib/libc.so.6
#5  0x0000000000415d33 in ?? ()
#6  0x00000000004137ef in ?? ()
#7  0x0000000000412af3 in ?? ()
#8  0x0000000000413933 in ?? ()
#9  0x0000000000412af3 in ?? ()
#10 0x0000000000412d81 in ?? ()
#11 0x0000000000414af6 in ?? ()
#12 0x00007f913f99d1a6 in __libc_start_main () from /lib/libc.so.6
#13 0x0000000000402839 in ?? ()
#14 0x00007fff487544f8 in ?? ()
#15 0x000000000000001c in ?? ()
#16 0x0000000000000005 in ?? ()
#17 0x00007fff48755f10 in ?? ()
#18 0x00007fff48755f17 in ?? ()
#19 0x00007fff48755f1a in ?? ()
#20 0x00007fff48755f1f in ?? ()
#21 0x00007fff48755f22 in ?? ()
#22 0x0000000000000000 in ?? ()
(gdb)

Same thing with 2.10.1:

read(3, "IN\201\244\1\2\0\1\0\0\0\0\0\0\0\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\7G"..., 16384) = 16384
lseek(3, 704512, SEEK_SET)              = 704512
read(3, "IN\201\244\1\2\0\1\0\0\0\0\0\0\0\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\21H"..., 16384) = 16384
lseek(3, 745472, SEEK_SET)              = 745472
read(3, "\177ELF\2\1\1\0\0\0\0\0\0\0\0\0\3\0>\0\1\0\0\0P\34\0\0\0\0\0\0@"..., 16384) = 16384
mmap(NULL, 421888, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f4de5c60000
munmap(0x7f4de5c60000, 421888)          = 0
mmap(NULL, 475136, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f4de5c53000
munmap(0x7f4de5c53000, 475136)          = 0
brk(0x195b000)                          = 0x195b000
mmap(NULL, 76083200, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f4de08b9000
munmap(0x7f4de08b9000, 76083200)        = 0
mmap(NULL, 21008228352, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f4900e46000
munmap(0x7f4900e46000, 21008228352)     = 0
mmap(NULL, 23702343680, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f48604f7000
munmap(0x7f48604f7000, 23702343680)     = 0
lseek(3, 835584, SEEK_SET)              = 835584
read(3, "IN\201\244\1\2\0\1\0\0\0\0\0\0\0\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\4H"..., 16384) = 16384
lseek(3, 1015808, SEEK_SET)             = 1015808
read(3, "IN\201\244\1\2\0\1\0\0\0\0\0\0\0\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\2H"..., 16384) = 16384
lseek(3, 1597440, SEEK_SET)             = 1597440
read(3, "IN\201\244\1\2\0\1\0\0\0\0\0\0\0\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\2H"..., 16384) = 16384
lseek(3, 1613824, SEEK_SET)             = 1613824
read(3, "IN\201\244\1\2\0\1\0\0\0\0\0\0\0\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\2H"..., 16384) = 16384
lseek(3, 1638400, SEEK_SET)             = 1638400
read(3, "IN\241\377\1\1\0\1\0\0\0\0\0\0\0\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\5I"..., 16384) = 16384
lseek(3, 1851392, SEEK_SET)             = 1851392
read(3, "IN\201\244\1\2\0\1\0\0\0\0\0\0\0\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\2H"..., 16384) = 16384
lseek(3, 1867776, SEEK_SET)             = 1867776
read(3, "IN\201\244\1\2\0\1\0\0\0\0\0\0\0\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\2H"..., 16384) = 16384
lseek(3, 1933312, SEEK_SET)             = 1933312
read(3, "INA\355\1\1\0\2\0\0\0\0\0\0\0\0\0\0\0\2\0\0\0\0\0\0\0\0\0\0\0KH"..., 16384) = 16384
lseek(3, 1966080, SEEK_SET)             = 1966080
read(3, "IN\201\264\1\2\0\1\0\0\0\0\0\0\0\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\2I"..., 16384) = 16384
lseek(3, 1982464, SEEK_SET)             = 1982464
read(3, "IN\201\264\1\2\0\1\0\0\0\0\0\0\0\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\2I"..., 16384) = 16384
lseek(3, 2260992, SEEK_SET)             = 2260992
read(3, "TZif2\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\2\0\0\0\2\0\0\0\0\0"..., 16384) = 16384
mmap(NULL, 31221911552, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = -1 ENOMEM (Cannot allocate memory)
brk(0x746873000)                        = 0x195b000
mmap(NULL, 31222042624, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = -1 ENOMEM (Cannot allocate memory)
rt_sigprocmask(SIG_BLOCK, [INT], NULL, 8) = 0
fstat(1, {st_mode=S_IFCHR|0620, st_rdev=makedev(136, 23), ...}) = 0
mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f4de5cdb000
write(1, "xfs_db: out of memory\n"..., 22xfs_db: out of memory
) = 22
rt_sigprocmask(SIG_UNBLOCK, [INT], NULL, 8) = 0
exit_group(4)                           = ?

xfs_db version 2.10.1

Justin.

Re: xfs_db: out of memory (2.9.8)

[Eric Sandeen]

Justin Piszcz wrote:
> 
> On Sat, 25 Oct 2008, Eric Sandeen wrote:
> 
>> Justin Piszcz wrote:
>>> For the root partition I get "out of memory" when I try to check for
>>> fragmentation.
>>>
>>> # xfs_db -c frag -f /dev/md2
>>> xfs_db: out of memory
>> "-f" means "stored in a file"
>>
>>> # mount | grep "/dev/md2"
>>> /dev/md2 on / type xfs (rw,noatime)
>> It looks like you're cheating your way around:
>>
>> xfs_db: /dev/md2 contains a mounted filesystem
>>
>> fatal error -- couldn't initialize XFS library
>>
>> I'm not sure if that matters in this respect, though.
>>
>> You could use gdb to find out which allocation is failing.
>>
>>> # xfs_db -V
>>> xfs_db version 2.9.8
>>>
>>> Any idea?
>> Don't worry so much about that fragmentation stat? :)
>>
>> -Eric
>>
>>> Justin.
>>>
>>>
> 
> .. but it used to work :(

if you can find out when it changed, maybe it can change back.

> Loaded symbols for /lib/libgcc_s.so.1
> Error while mapping shared library sections:
> ease: No such file or directory.
> (no debugging symbols found)

... so the output isn't so useful here

> mmap(NULL, 31221911552, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = -1 ENOMEM (Cannot allocate memory)

32G...

-Eric