View previous topic :: View next topic |
Author |
Message |
wbecker Tux's lil' helper
Joined: 16 May 2005 Posts: 95 Location: Melbourne, Australia
|
Posted: Mon Jan 18, 2010 4:50 pm Post subject: [solved] Can't load more than about 256 MB of memory/process |
|
|
It's really bizarre. I have several amd64 systems running but on one particular one, I find that as soon as an application goes over about 256MB ram it either starts using swap or (if I turn swap off) it just crashes.
I have 4GB of memory on this machine, so it oughtn't be a memory issue.
I thought it was just eclipse at the start - so it was a Java issue but I have fiddled with eclipserc to no avail. I then found that plasma and vmware and all sorts of things trigger it.
Is there some funny kernel setting that I could have on which I shouldn't?
I dont use any funny use flags, just:
Code: | USE="-semantic-desktop mmx sse sse2 -gtk xinerama truetype type1 cleartype corefonts zip kde -gnome nsplugin -gstreamer -eds" |
Is there anything I can do?
-------------
PS: Looking at dmesg after a crash I get:
Code: | vmware-vmx invoked oom-killer: gfp_mask=0x201da, order=0, oomkilladj=0
vmware-vmx cpuset=/ mems_allowed=0
Pid: 4856, comm: vmware-vmx Tainted: P 2.6.31-gentoo-r6 #5
Call Trace:
[<ffffffff810909e6>] ? oom_kill_process+0x91/0x23d
[<ffffffff8102c371>] ? read_tsc+0xa/0x20
[<ffffffff81090d52>] ? badness+0x1a2/0x1e6
[<ffffffff81090dd2>] ? __out_of_memory+0x3c/0x142
[<ffffffff8109101a>] ? out_of_memory+0x142/0x174
[<ffffffff810c85c3>] ? do_sys_poll+0x30e/0x389
[<ffffffff810942b0>] ? __alloc_pages_nodemask+0x45e/0x53a
[<ffffffff812be6a0>] ? scsi_request_fn+0x369/0x440
[<ffffffff81095c20>] ? __do_page_cache_readahead+0x9b/0x1a4
[<ffffffff81095f37>] ? ondemand_readahead+0x11f/0x1a7
[<ffffffff8108fb30>] ? generic_file_aio_read+0x1ff/0x527
[<ffffffff810bbb25>] ? do_sync_read+0xce/0x113
[<ffffffff810a16be>] ? handle_mm_fault+0x2eb/0xa57
[<ffffffff81069286>] ? autoremove_wake_function+0x0/0x2e
[<ffffffff8106243b>] ? group_send_sig_info+0x18/0x6f
[<ffffffff810bc480>] ? vfs_read+0xa2/0xdb
[<ffffffff810bc510>] ? sys_pread64+0x57/0x77
[<ffffffff81026a82>] ? system_call_fastpath+0x16/0x1b
Mem-Info:
Node 0 DMA per-cpu:
CPU 0: hi: 0, btch: 1 usd: 0
CPU 1: hi: 0, btch: 1 usd: 0
CPU 2: hi: 0, btch: 1 usd: 0
CPU 3: hi: 0, btch: 1 usd: 0
Node 0 DMA32 per-cpu:
CPU 0: hi: 186, btch: 31 usd: 49
CPU 1: hi: 186, btch: 31 usd: 54
CPU 2: hi: 186, btch: 31 usd: 97
CPU 3: hi: 186, btch: 31 usd: 166
Node 0 Normal per-cpu:
CPU 0: hi: 186, btch: 31 usd: 47
CPU 1: hi: 186, btch: 31 usd: 82
CPU 2: hi: 186, btch: 31 usd: 93
CPU 3: hi: 186, btch: 31 usd: 160
Active_anon:162583 active_file:253064 inactive_anon:65158
inactive_file:480675 unevictable:1 dirty:95550 writeback:0 unstable:0
free:6721 slab:16920 mapped:52109 pagetables:5483 bounce:0
Node 0 DMA free:15912kB min:28kB low:32kB high:40kB active_anon:0kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB present:15344kB pages_scanned:0 all_unreclaimable? yes
lowmem_reserve[]: 0 3255 4013 4013
Node 0 DMA32 free:9504kB min:6564kB low:8204kB high:9844kB active_anon:512144kB inactive_anon:122332kB active_file:860340kB inactive_file:1674356kB unevictable:0kB present:3334048kB pages_scanned:5939272 all_unreclaimable? yes
lowmem_reserve[]: 0 0 757 757
Node 0 Normal free:1468kB min:1524kB low:1904kB high:2284kB active_anon:138188kB inactive_anon:138300kB active_file:151916kB inactive_file:248344kB unevictable:4kB present:775680kB pages_scanned:1039457 all_unreclaimable? yes
lowmem_reserve[]: 0 0 0 0
Node 0 DMA: 4*4kB 3*8kB 4*16kB 2*32kB 4*64kB 1*128kB 2*256kB 1*512kB 2*1024kB 2*2048kB 2*4096kB = 15912kB
Node 0 DMA32: 962*4kB 5*8kB 3*16kB 6*32kB 8*64kB 2*128kB 4*256kB 1*512kB 1*1024kB 1*2048kB 0*4096kB = 9504kB
Node 0 Normal: 45*4kB 161*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1468kB
739827 total pagecache pages
1048576 pages RAM
33785 pages reserved
838881 pages shared
305159 pages non-shared
Out of memory: kill process 11152 (kdeinit4) score 1761657 or a child
Killed process 11153 (klauncher)
vmware-hostd invoked oom-killer: gfp_mask=0xd0, order=0, oomkilladj=0
vmware-hostd cpuset=/ mems_allowed=0
Pid: 4415, comm: vmware-hostd Tainted: P 2.6.31-gentoo-r6 #5
Call Trace:
[<ffffffff810909e6>] ? oom_kill_process+0x91/0x23d
[<ffffffff8102c371>] ? read_tsc+0xa/0x20
[<ffffffff81090d52>] ? badness+0x1a2/0x1e6
[<ffffffff81090dd2>] ? __out_of_memory+0x3c/0x142
[<ffffffff8109101a>] ? out_of_memory+0x142/0x174
[<ffffffff814c8fae>] ? thread_return+0x41/0xd2
[<ffffffff810942b0>] ? __alloc_pages_nodemask+0x45e/0x53a
[<ffffffff8106bb91>] ? hrtimer_cancel+0xc/0x16
[<ffffffff81093625>] ? __get_free_pages+0x9/0x46
[<ffffffff810c92ad>] ? __pollwait+0x5b/0xd6
[<ffffffff8145df9b>] ? sock_poll_wait+0x13/0x18
[<ffffffff8145dfb1>] ? unix_poll+0x11/0x90
[<ffffffff810c84b9>] ? do_sys_poll+0x204/0x389
[<ffffffff810c9252>] ? __pollwait+0x0/0xd6
[<ffffffff810c9328>] ? pollwake+0x0/0x58
[<ffffffff810c9328>] ? pollwake+0x0/0x58
[<ffffffff810c9328>] ? pollwake+0x0/0x58
[<ffffffff810c9328>] ? pollwake+0x0/0x58
[<ffffffff810c9328>] ? pollwake+0x0/0x58
[<ffffffff810c9328>] ? pollwake+0x0/0x58
[<ffffffff810c9328>] ? pollwake+0x0/0x58
[<ffffffff810c9328>] ? pollwake+0x0/0x58
[<ffffffff810c9328>] ? pollwake+0x0/0x58
[<ffffffff8106f5d5>] ? getnstimeofday+0x55/0xaf
[<ffffffff8106bfdc>] ? ktime_get_ts+0x21/0x49
[<ffffffff810c87cb>] ? sys_poll+0x47/0xb2
[<ffffffff81026a82>] ? system_call_fastpath+0x16/0x1b
Mem-Info:
Node 0 DMA per-cpu:
CPU 0: hi: 0, btch: 1 usd: 0
CPU 1: hi: 0, btch: 1 usd: 0
CPU 2: hi: 0, btch: 1 usd: 0
CPU 3: hi: 0, btch: 1 usd: 0
Node 0 DMA32 per-cpu:
CPU 0: hi: 186, btch: 31 usd: 57
CPU 1: hi: 186, btch: 31 usd: 186
CPU 2: hi: 186, btch: 31 usd: 98
CPU 3: hi: 186, btch: 31 usd: 167
Node 0 Normal per-cpu:
CPU 0: hi: 186, btch: 31 usd: 160
CPU 1: hi: 186, btch: 31 usd: 79
CPU 2: hi: 186, btch: 31 usd: 104
CPU 3: hi: 186, btch: 31 usd: 130
Active_anon:162455 active_file:253012 inactive_anon:65157
inactive_file:480330 unevictable:1 dirty:95556 writeback:0 unstable:0
free:6721 slab:16920 mapped:52038 pagetables:5399 bounce:0
Node 0 DMA free:15912kB min:28kB low:32kB high:40kB active_anon:0kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB present:15344kB pages_scanned:0 all_unreclaimable? yes
lowmem_reserve[]: 0 3255 4013 4013
Node 0 DMA32 free:9504kB min:6564kB low:8204kB high:9844kB active_anon:511580kB inactive_anon:122332kB active_file:860164kB inactive_file:1672988kB unevictable:0kB present:3334048kB pages_scanned:4899850 all_unreclaimable? yes
lowmem_reserve[]: 0 0 757 757
Node 0 Normal free:1468kB min:1524kB low:1904kB high:2284kB active_anon:138240kB inactive_anon:138296kB active_file:151884kB inactive_file:248332kB unevictable:4kB present:775680kB pages_scanned:1131616 all_unreclaimable? yes
lowmem_reserve[]: 0 0 0 0
Node 0 DMA: 4*4kB 3*8kB 4*16kB 2*32kB 4*64kB 1*128kB 2*256kB 1*512kB 2*1024kB 2*2048kB 2*4096kB = 15912kB
Node 0 DMA32: 980*4kB 2*8kB 0*16kB 6*32kB 8*64kB 2*128kB 4*256kB 1*512kB 1*1024kB 1*2048kB 0*4096kB = 9504kB
Node 0 Normal: 59*4kB 154*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1468kB
739830 total pagecache pages
1048576 pages RAM
33785 pages reserved
836521 pages shared
304945 pages non-shared
Out of memory: kill process 11152 (kdeinit4) score 1736474 or a child
Killed process 11179 (ksmserver) |
Last edited by wbecker on Wed Jan 20, 2010 10:21 am; edited 1 time in total |
|
Back to top |
|
|
aCOSwt Bodhisattva
Joined: 19 Oct 2007 Posts: 2537 Location: Hilbert space
|
Posted: Mon Jan 18, 2010 5:47 pm Post subject: |
|
|
Could you post the output of
|
|
Back to top |
|
|
wbecker Tux's lil' helper
Joined: 16 May 2005 Posts: 95 Location: Melbourne, Australia
|
Posted: Tue Jan 19, 2010 9:06 am Post subject: |
|
|
It is this:
Code: | core file size (blocks, -c) 0
data seg size (kbytes, -d) unlimited
scheduling priority (-e) 0
file size (blocks, -f) unlimited
pending signals (-i) 38912
max locked memory (kbytes, -l) 64
max memory size (kbytes, -m) unlimited
open files (-n) 1024
pipe size (512 bytes, -p) 8
POSIX message queues (bytes, -q) 819200
real-time priority (-r) 0
stack size (kbytes, -s) 8192
cpu time (seconds, -t) unlimited
max user processes (-u) 38912
virtual memory (kbytes, -v) unlimited
file locks (-x) unlimited
|
I tried it on another computer which works fine and it is mostly the same... |
|
Back to top |
|
|
aCOSwt Bodhisattva
Joined: 19 Oct 2007 Posts: 2537 Location: Hilbert space
|
Posted: Tue Jan 19, 2010 9:29 am Post subject: |
|
|
And the output of :
Code: | cat /proc/sys/vm/swappiness |
|
|
Back to top |
|
|
wbecker Tux's lil' helper
Joined: 16 May 2005 Posts: 95 Location: Melbourne, Australia
|
Posted: Tue Jan 19, 2010 9:30 am Post subject: |
|
|
I havent changed it manually - it is currently on 60 |
|
Back to top |
|
|
LesCoke n00b
Joined: 02 Jun 2008 Posts: 48 Location: Denton, Tx
|
Posted: Tue Jan 19, 2010 9:32 am Post subject: |
|
|
Check how you have your kernel memory model setup. Specifically check CONFIG_NOHIGHMEM, CONFIG_HIGMEM4G and CONFIG_HIGHMEM64G. I'm pulling these from an x86 2.6.23 kernel, amd64 and later kernels may have different names. Having 4G of RAM uses the whole 32bit address space leaving no place for the kernel to page VM into, thus these control which PAE model the kernel uses.
But I'm puzzled why the kernel limited the process to only 256M.
Les |
|
Back to top |
|
|
wbecker Tux's lil' helper
Joined: 16 May 2005 Posts: 95 Location: Melbourne, Australia
|
Posted: Tue Jan 19, 2010 9:41 am Post subject: |
|
|
That's strange...
If I go make menuconfig, and do a search it appears and tells me that I don't have HIGHMEM selected, but if I go to the place in Processor type and Features, then it is not there... Nor is there any thing to do with HIGHMEM in the .config file.
Has this changed or is it set by default? |
|
Back to top |
|
|
aCOSwt Bodhisattva
Joined: 19 Oct 2007 Posts: 2537 Location: Hilbert space
|
Posted: Tue Jan 19, 2010 9:44 am Post subject: |
|
|
In the VMWare Server Console, what is the setting for the tweak memory/swap use ? |
|
Back to top |
|
|
wbecker Tux's lil' helper
Joined: 16 May 2005 Posts: 95 Location: Melbourne, Australia
|
Posted: Tue Jan 19, 2010 9:48 am Post subject: |
|
|
Oh....
I have reserved 3000MB for VMware. I figured that it would only prevent other processes from using it if it needed it... But it could be that it just prevents anyone else from using that 3Gig.
I have moved it down to 1000MB - I'll see if that helps.
Thanks! |
|
Back to top |
|
|
aCOSwt Bodhisattva
Joined: 19 Oct 2007 Posts: 2537 Location: Hilbert space
|
Posted: Tue Jan 19, 2010 9:49 am Post subject: |
|
|
wbecker wrote: | Oh....
I have reserved 3000MB for VMware. |
|
|
Back to top |
|
|
LesCoke n00b
Joined: 02 Jun 2008 Posts: 48 Location: Denton, Tx
|
Posted: Tue Jan 19, 2010 11:10 pm Post subject: |
|
|
I checked my older 2.6.19 x86_64 system and found that it was not an option in that kernel; only found FLATMEM and SPARSEMEM. My ulimits are roughly the same and your values look good. I haven't updated this older Sempron64 system since it still has hard drive failures I haven't fixed yet.
Are you certain swap is setup?
I'm currently out of ideas.
Les |
|
Back to top |
|
|
wbecker Tux's lil' helper
Joined: 16 May 2005 Posts: 95 Location: Melbourne, Australia
|
Posted: Wed Jan 20, 2010 10:19 am Post subject: |
|
|
I'm pretty sure it was my VMware setting. It's all working now. |
|
Back to top |
|
|
|