[Users] vzubc, memory, and buffers
Keith Keller
kkeller at wombat.san-francisco.ca.us
Thu Jul 5 14:05:11 EDT 2012
Hi again users,
I wanted to ask a quick question about the vzubc utility and buffers
inside a container. I have a container which has a lot of disk buffers
currently being stored in memory:
$ free
total used free shared buffers cached
Mem: 4194304 4116528 77776 0 0 3591588
-/+ buffers/cache: 524940 3669364
Swap: 524288 0 524288
This is IME a perfectly normal and acceptable situation (and the
container is behaving normally).
However, vzubc thinks that this is a warning situation:
# vzubc -q -r
----------------------------------------------------------------
CT 1018 | HELD Bar% Lim%| MAXH Bar% Lim%| BAR | LIM |+FAIL
-------------+---------------+---------------+-----+-----+------
physpages|3.93G - 98%| 4G - 100%| - | 4G| -
----------------------------------------------------------------
What do people normally do in this situation? Would you simply ignore
warnings on physpages, or actively check the free buffers situation in
that CT?
In what might be a related question, I saw that vzcfgvalidate reports a
somewhat strange error in this CT:
# vzcfgvalidate 1018.conf
Warning: swappages.lim should be <= 0 (currently, 131072)
# grep SWAP 1018.conf
SWAPPAGES="0:512M"
But isn't swappages.lim set to 512M here? And why should it be set to
less than or equal to 0? And why does vzcfgvalidate think it's some
number not 0 or 512M?
If it's helpful, here's my list of packages (this is a CentOS 6 box):
vzctl.x86_64 3.3-1 @openvz-utils
vzctl-lib.x86_64 3.3-1 @openvz-utils
vzkernel.x86_64 2.6.32-042stab055.10 @openvz-kernel-rhel6
vzkernel-firmware.noarch 2.6.32-042stab055.10 @openvz-kernel-rhel6
vzquota.x86_64 3.0.12-1 @openvz-utils
And kernel (yes, I know it is not the latest):
# uname -a
Linux xxxxxxxxxx 2.6.32-042stab055.10 #1 SMP Thu May 10 15:38:32 MSD 2012 x86_64 x86_64 x86_64 GNU/Linux
Thanks for any advice you have!
--keith
--
kkeller at wombat.san-francisco.ca.us
More information about the Users
mailing list