| ENOMEM errors [message #30796] |
Sat, 07 June 2008 06:47  |
pzttilde
Messages: 3
Registered: June 2008
|
Junior Member |
|
|
Hey there, inside my machines I've been getting ENOMEM errors and I'm not quite sure how to handle it. The machine's 'host' has 8G of memory of which it caches everything but what is used at that point as soon as I start any container. Not sure if this behavior is intended or faulty. Inside the container only 256M seems to be available.
It's an Ubuntu Hardy install;
Linux merlin.*** 2.6.24-19-openvz #1 SMP Thu Jun 5 13:07:13 CEST 2008 x86_64 GNU/Linux
This is the host:
# free -m
total used free shared buffers cached
Mem: 7981 7933 47 0 0 5850
-/+ buffers/cache: 2083 5897
Swap: 22888 0 22888
Inside container (the only one running at this point):
# free -m
total used free shared buffers cached
Mem: 256 60 195 0 0 0
# cat /proc/user_beancounters | grep privv
privvmpages 15638 200464 73728 80362 486040
# sysctl -a | grep overcommit
vm.overcommit_memory = 0
vm.overcommit_ratio = 50
vm.nr_overcommit_hugepages = 0
# ulimit -a
core file size (blocks, -c) 0
data seg size (kbytes, -d) unlimited
scheduling priority (-e) 0
file size (blocks, -f) unlimited
pending signals (-i) 71680
max locked memory (kbytes, -l) 32
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) 71680
virtual memory (kbytes, -v) unlimited
file locks (-x) unlimited
Feel like I'm just doing something essential wrong or that I've stumbled upon a bug. Any advice would be appreciated.
|
|
|
|
|
|
| Re: ENOMEM errors [message #30832 is a reply to message #30804] |
Sun, 08 June 2008 07:50  |
pzttilde
Messages: 3
Registered: June 2008
|
Junior Member |
|
|
| maratrus wrote on Sat, 07 June 2008 05:48 |
Hello,
excuse me but I don't understand what's your problem?
Do you want to increase privvmpages parameter or what?
|
Well, thanks to this error sometimes the vm just crashes because the host is out of memory. And less bad (but still very annoying.), inside the vm applications have the same issue.
For example:
# vzctl enter 10
Unable to fork: Cannot allocate memory
Are my values strange? I haven't really changed them much. And is it normal for the host to cache all the memory like that?
[Updated on: Sun, 08 June 2008 08:04] Report message to a moderator |
|
|
|
|
|
| Re: ENOMEM errors [message #31009 is a reply to message #30836] |
Thu, 12 June 2008 18:32 |
pzttilde
Messages: 3
Registered: June 2008
|
Junior Member |
|
|
| maratrus wrote on Sun, 08 June 2008 08:56 |
Try to increase the value of privvmpages.
For example, you can do it via vzctl (man vzctl):
vzctl set CT_ID --privvmpages [barrier]:[limit] --save
|
That was indeed the issue, thanks. Is it normal for 'free' to not understand the actual memory available to the system? It always shows 256 for me, even though a lot more is actually used.
On another note, I'm quite impressed by openvz!
[Updated on: Thu, 12 June 2008 18:34] Report message to a moderator |
|
|
|
OpenVZ Forum
Members
Search
Help
Register
Login
Home
