ia64/linux-2.6.18-xen.hg

view Documentation/networking/proc_net_tcp.txt @ 897:329ea0ccb344

balloon: try harder to balloon up under memory pressure.

Currently if the balloon driver is unable to increase the guest's
reservation it assumes the failure was due to reaching its full
allocation, gives up on the ballooning operation and records the limit
it reached as the "hard limit". The driver will not try again until
the target is set again (even to the same value).

However it is possible that ballooning has in fact failed due to
memory pressure in the host and therefore it is desirable to keep
attempting to reach the target in case memory becomes available. The
most likely scenario is that some guests are ballooning down while
others are ballooning up and therefore there is temporary memory
pressure while things stabilise. You would not expect a well behaved
toolstack to ask a domain to balloon to more than its allocation nor
would you expect it to deliberately over-commit memory by setting
balloon targets which exceed the total host memory.

This patch drops the concept of a hard limit and causes the balloon
driver to retry increasing the reservation on a timer in the same
manner as when decreasing the reservation.

Also if we partially succeed in increasing the reservation
(i.e. receive less pages than we asked for) then we may as well keep
those pages rather than returning them to Xen.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
author Keir Fraser <keir.fraser@citrix.com>
date Fri Jun 05 14:01:20 2009 +0100 (2009-06-05)
parents 831230e53067
children
line source
1 This document describes the interfaces /proc/net/tcp and /proc/net/tcp6.
3 These /proc interfaces provide information about currently active TCP
4 connections, and are implemented by tcp_get_info() in net/ipv4/tcp_ipv4.c and
5 tcp6_get_info() in net/ipv6/tcp_ipv6.c, respectively.
7 It will first list all listening TCP sockets, and next list all established
8 TCP connections. A typical entry of /proc/net/tcp would look like this (split
9 up into 3 parts because of the length of the line):
11 46: 010310AC:9C4C 030310AC:1770 01
12 | | | | | |--> connection state
13 | | | | |------> remote TCP port number
14 | | | |-------------> remote IPv4 address
15 | | |--------------------> local TCP port number
16 | |---------------------------> local IPv4 address
17 |----------------------------------> number of entry
19 00000150:00000000 01:00000019 00000000
20 | | | | |--> number of unrecovered RTO timeouts
21 | | | |----------> number of jiffies until timer expires
22 | | |----------------> timer_active (see below)
23 | |----------------------> receive-queue
24 |-------------------------------> transmit-queue
26 1000 0 54165785 4 cd1e6040 25 4 27 3 -1
27 | | | | | | | | | |--> slow start size threshold,
28 | | | | | | | | | or -1 if the treshold
29 | | | | | | | | | is >= 0xFFFF
30 | | | | | | | | |----> sending congestion window
31 | | | | | | | |-------> (ack.quick<<1)|ack.pingpong
32 | | | | | | |---------> Predicted tick of soft clock
33 | | | | | | (delayed ACK control data)
34 | | | | | |------------> retransmit timeout
35 | | | | |------------------> location of socket in memory
36 | | | |-----------------------> socket reference count
37 | | |-----------------------------> inode
38 | |----------------------------------> unanswered 0-window probes
39 |---------------------------------------------> uid
41 timer_active:
42 0 no timer is pending
43 1 retransmit-timer is pending
44 2 another timer (e.g. delayed ack or keepalive) is pending
45 3 this is a socket in TIME_WAIT state. Not all fields will contain
46 data (or even exist)
47 4 zero window probe timer is pending