[lxc-devel] LXC-Cgroup memory controller
Hi Guys I wanna know if it possible to set up a memory controller for each container..if so..how can I do? any examples? Thanks Jorge E. Espada -- This SF.net Dev2Dev email is sponsored by: Show off your parallel programming skills. Enter the Intel(R) Threading Challenge 2010. http://p.sf.net/sfu/intel-thread-sfd___ Lxc-devel mailing list Lxc-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lxc-devel
Re: [lxc-devel] LXC-Cgroup memory controller
Thank you Daniel, but in order to do this I must have kernel support , right? dev2:~# lxc-checkconfig | grep memory Cgroup memory controller: disabled I think I need this property enabled Thanks in advance Jorge E. Espada Phone: +54 9 341 5692435 On Tue, Sep 7, 2010 at 3:00 PM, Daniel Lezcano wrote: > On 09/07/2010 05:22 PM, jorge espada wrote: > >> Hi Guys I wanna know if it possible to set up a memory controller for each >> container..if so..how can I do? any examples? >> Thanks >> >> > > When you start a container, there is a cgroup created automatically where > all the processes of the container will belong. > > In order to modify a cgroup subsystem value use the lxc-cgroup -n > > > For example : lxc-cgroup -n foo memory.max_usage_in_bytes 268435456 > > Or specify it in the configuration file to automatically set it up when the > container starts. > > lxc.cgroup.memory.max_usage_in_bytes = 268435456 > > -- Daniel > -- This SF.net Dev2Dev email is sponsored by: Show off your parallel programming skills. Enter the Intel(R) Threading Challenge 2010. http://p.sf.net/sfu/intel-thread-sfd___ Lxc-devel mailing list Lxc-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lxc-devel
[lxc-devel] lxc.network.type = macvlan
Hi, I have a lxc container running on debian lenny (kernel 2.6.32) with the following conf: lxc.network.type = macvlan lxc.network.flags = up lxc.network.link = vlan310 lxc.network.name = vlan310 lxc.network.macvlan.mode = vepa lxc.network.hwaddr = 06:00:0a:1f:00:46 lxc.network.ipv4 = 10.31.0.70/24 # network2 lxc.network.type = macvlan lxc.network.flags = up lxc.network.link = vlan400 lxc.network.name = vlan400 lxc.network.macvlan.mode = vepa lxc.network.hwaddr = 06:00:0a:28:00:46 lxc.network.ipv4 = 10.40.0.70/23 is the macvlan mode vepa supported for that kernel or I need a newer one? Because I have weird problems with that conf, some times I can ping the container from another host but some times not...is random Jorge E. Espada -- Beautiful is writing same markup. Internet Explorer 9 supports standards for HTML5, CSS3, SVG 1.1, ECMAScript5, and DOM L2 & L3. Spend less time writing and rewriting code and more time creating great experiences on the web. Be a part of the beta today. http://p.sf.net/sfu/beautyoftheweb___ Lxc-devel mailing list Lxc-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lxc-devel
Re: [lxc-devel] lxc.network.type = macvlan
The error was ICMP "no route to host", but now i change the conf and instead of vepa I used bridge...after destroy and create the containers with the new conf, seems to be working as expected for a while, but from time to time "the container became unresponsive".. no ping (100% packet lost). The curious thing is, if I login into the host and then to the container (lxc-console) and ping another container or host the " unresponsive container " become alive again (ping OK)so I'm a bit lost.. Jorge E. Espada On Wed, Oct 13, 2010 at 2:17 AM, Daniel Lezcano wrote: > On 10/12/2010 10:29 PM, jorge espada wrote: > >> Hi, I have a lxc container running on debian lenny (kernel 2.6.32) with >> the >> following conf: >> >> lxc.network.type = macvlan >> lxc.network.flags = up >> lxc.network.link = vlan310 >> lxc.network.name = vlan310 >> lxc.network.macvlan.mode = vepa >> lxc.network.hwaddr = 06:00:0a:1f:00:46 >> lxc.network.ipv4 = 10.31.0.70/24 >> >> # network2 >> lxc.network.type = macvlan >> lxc.network.flags = up >> lxc.network.link = vlan400 >> lxc.network.name = vlan400 >> lxc.network.macvlan.mode = vepa >> lxc.network.hwaddr = 06:00:0a:28:00:46 >> lxc.network.ipv4 = 10.40.0.70/23 >> >> >> is the macvlan mode vepa supported for that kernel or I need a newer one? >> > > You need a 2.6.33 kernel version. > > > Because I have weird problems with that conf, some times I can ping the >> container from another host but some times not...is random >> > > I am not sure it is related. What happens when you can't ping the host ? > What is the error ? ICMP "no route to host" ? or just the packet are lost ? > -- Beautiful is writing same markup. Internet Explorer 9 supports standards for HTML5, CSS3, SVG 1.1, ECMAScript5, and DOM L2 & L3. Spend less time writing and rewriting code and more time creating great experiences on the web. Be a part of the beta today. http://p.sf.net/sfu/beautyoftheweb___ Lxc-devel mailing list Lxc-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lxc-devel
Re: [lxc-devel] lxc.network.type = macvlan
No, from the host: Oct 10 06:25:01 srvc15 kernel: imklog 4.4.2, log source = /proc/kmsg started. Oct 10 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] (re)start Oct 11 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] rsyslogd was HUPed, type 'restart'. Oct 11 06:25:01 srvc15 kernel: imklog 4.4.2, log source = /proc/kmsg started. Oct 11 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] (re)start Oct 12 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] rsyslogd was HUPed, type 'restart'. Oct 12 06:25:01 srvc15 kernel: Kernel logging (proc) stopped. Oct 12 06:25:01 srvc15 kernel: imklog 4.4.2, log source = /proc/kmsg started. Oct 12 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] (re)start Oct 13 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] rsyslogd was HUPed, type 'restart'. Oct 13 06:25:01 srvc15 kernel: Kernel logging (proc) stopped. Oct 13 06:25:01 srvc15 kernel: imklog 4.4.2, log source = /proc/kmsg started. Oct 13 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] (re)start >From the container: Oct 10 06:25:01 srvc15c0 kernel: imklog 3.18.6, log source = /proc/kmsg started. Oct 10 06:25:01 srvc15c0 rsyslogd: [origin software="rsyslogd" swVersion="3.18.6" x-pid="299" x-info="http://www.rsyslog.com";] restart Oct 11 06:25:01 srvc15c0 kernel: imklog 3.18.6, log source = /proc/kmsg started. Oct 11 06:25:01 srvc15c0 rsyslogd: [origin software="rsyslogd" swVersion="3.18.6" x-pid="299" x-info="http://www.rsyslog.com";] restart Oct 12 06:25:01 srvc15c0 kernel: imklog 3.18.6, log source = /proc/kmsg started. Oct 12 06:25:01 srvc15c0 rsyslogd: [origin software="rsyslogd" swVersion="3.18.6" x-pid="299" x-info="http://www.rsyslog.com";] restart Oct 13 06:25:01 srvc15c0 kernel: imklog 3.18.6, log source = /proc/kmsg started. Oct 13 06:25:01 srvc15c0 rsyslogd: [origin software="rsyslogd" swVersion="3.18.6" x-pid="299" x-info="http://www.rsyslog.com";] restart That's all :( Jorge E. Espada On Wed, Oct 13, 2010 at 9:45 AM, Daniel Lezcano wrote: > On 10/13/2010 02:42 PM, jorge espada wrote: > >> The error was ICMP "no route to host", but now i change the conf and >> instead of vepa I used bridge...after destroy and create the containers >> with >> the new conf, seems to be working as expected for a while, but from time >> to >> time "the container became unresponsive".. no ping (100% packet lost). The >> curious thing is, if I login into the host and then to the container >> (lxc-console) and ping another container or host the " unresponsive >> container " become alive again (ping OK)so I'm a bit lost.. >> > > Is there any trace in /var/log/messages ? > -- Beautiful is writing same markup. Internet Explorer 9 supports standards for HTML5, CSS3, SVG 1.1, ECMAScript5, and DOM L2 & L3. Spend less time writing and rewriting code and more time creating great experiences on the web. Be a part of the beta today. http://p.sf.net/sfu/beautyoftheweb___ Lxc-devel mailing list Lxc-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lxc-devel
Re: [lxc-devel] lxc.network.type = macvlan
As extra information I have configured a bonding in the host(where the container is running) and the vlans over it: auto bond0 iface bond0 inet manual up /sbin/ifconfig bond0 up down /sbin/ifconfig bond0 down slaves eth0 eth1 bond_mode balance-xor bond_miimon 100 bond_updelay 200 bond_downdelay 200 auto vlan310 iface vlan310 inet static address 10.31.0.151 netmask 255.255.255.0 network 10.31.0.0 broadcast 10.31.0.255 gateway 10.31.0.1 vlan_raw_device bond0 Jorge E. Espada On Wed, Oct 13, 2010 at 10:06 AM, jorge espada wrote: > No, from the host: > > Oct 10 06:25:01 srvc15 kernel: imklog 4.4.2, log source = /proc/kmsg > started. > Oct 10 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" > swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] (re)start > Oct 11 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" > swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] rsyslogd > was HUPed, type 'restart'. > Oct 11 06:25:01 srvc15 kernel: imklog 4.4.2, log source = /proc/kmsg > started. > Oct 11 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" > swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] (re)start > Oct 12 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" > swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] rsyslogd > was HUPed, type 'restart'. > Oct 12 06:25:01 srvc15 kernel: Kernel logging (proc) stopped. > Oct 12 06:25:01 srvc15 kernel: imklog 4.4.2, log source = /proc/kmsg > started. > Oct 12 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" > swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] (re)start > Oct 13 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" > swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] rsyslogd > was HUPed, type 'restart'. > Oct 13 06:25:01 srvc15 kernel: Kernel logging (proc) stopped. > Oct 13 06:25:01 srvc15 kernel: imklog 4.4.2, log source = /proc/kmsg > started. > Oct 13 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" > swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] (re)start > > From the container: > Oct 10 06:25:01 srvc15c0 kernel: imklog 3.18.6, log source = /proc/kmsg > started. > Oct 10 06:25:01 srvc15c0 rsyslogd: [origin software="rsyslogd" > swVersion="3.18.6" x-pid="299" x-info="http://www.rsyslog.com";] restart > Oct 11 06:25:01 srvc15c0 kernel: imklog 3.18.6, log source = /proc/kmsg > started. > Oct 11 06:25:01 srvc15c0 rsyslogd: [origin software="rsyslogd" > swVersion="3.18.6" x-pid="299" x-info="http://www.rsyslog.com";] restart > Oct 12 06:25:01 srvc15c0 kernel: imklog 3.18.6, log source = /proc/kmsg > started. > Oct 12 06:25:01 srvc15c0 rsyslogd: [origin software="rsyslogd" > swVersion="3.18.6" x-pid="299" x-info="http://www.rsyslog.com";] restart > Oct 13 06:25:01 srvc15c0 kernel: imklog 3.18.6, log source = /proc/kmsg > started. > Oct 13 06:25:01 srvc15c0 rsyslogd: [origin software="rsyslogd" > swVersion="3.18.6" x-pid="299" x-info="http://www.rsyslog.com";] restart > > That's all :( > Jorge E. Espada > > > > > On Wed, Oct 13, 2010 at 9:45 AM, Daniel Lezcano wrote: > >> On 10/13/2010 02:42 PM, jorge espada wrote: >> >>> The error was ICMP "no route to host", but now i change the conf and >>> instead of vepa I used bridge...after destroy and create the containers >>> with >>> the new conf, seems to be working as expected for a while, but from time >>> to >>> time "the container became unresponsive".. no ping (100% packet lost). >>> The >>> curious thing is, if I login into the host and then to the container >>> (lxc-console) and ping another container or host the " unresponsive >>> container " become alive again (ping OK)so I'm a bit lost.. >>> >> >> Is there any trace in /var/log/messages ? >> > > -- Beautiful is writing same markup. Internet Explorer 9 supports standards for HTML5, CSS3, SVG 1.1, ECMAScript5, and DOM L2 & L3. Spend less time writing and rewriting code and more time creating great experiences on the web. Be a part of the beta today. http://p.sf.net/sfu/beautyoftheweb___ Lxc-devel mailing list Lxc-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lxc-devel
Re: [lxc-devel] lxc.network.type = macvlan
My bad guys, it was a route problem...i have 3 vlans for each container and also 3 default gw in /etc/network/interfaces (copy and paste :S )...and no custom route...so... I modified the conf in /etc/network/interfaces and solved the problem. Thanks Jorge E. Espada On Thu, Oct 14, 2010 at 11:40 AM, jorge espada wrote: > As extra information I have configured a bonding in the host(where the > container is running) and the vlans over it: > auto bond0 > iface bond0 inet manual > up /sbin/ifconfig bond0 up > down /sbin/ifconfig bond0 down > slaves eth0 eth1 > bond_mode balance-xor > bond_miimon 100 > bond_updelay 200 > bond_downdelay 200 > > auto vlan310 > iface vlan310 inet static > address 10.31.0.151 > netmask 255.255.255.0 > network 10.31.0.0 > broadcast 10.31.0.255 > gateway 10.31.0.1 > vlan_raw_device bond0 > > > Jorge E. Espada > > > > > On Wed, Oct 13, 2010 at 10:06 AM, jorge espada wrote: > >> No, from the host: >> >> Oct 10 06:25:01 srvc15 kernel: imklog 4.4.2, log source = /proc/kmsg >> started. >> Oct 10 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" >> swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] (re)start >> Oct 11 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" >> swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] rsyslogd >> was HUPed, type 'restart'. >> Oct 11 06:25:01 srvc15 kernel: imklog 4.4.2, log source = /proc/kmsg >> started. >> Oct 11 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" >> swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] (re)start >> Oct 12 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" >> swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] rsyslogd >> was HUPed, type 'restart'. >> Oct 12 06:25:01 srvc15 kernel: Kernel logging (proc) stopped. >> Oct 12 06:25:01 srvc15 kernel: imklog 4.4.2, log source = /proc/kmsg >> started. >> Oct 12 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" >> swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] (re)start >> Oct 13 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" >> swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] rsyslogd >> was HUPed, type 'restart'. >> Oct 13 06:25:01 srvc15 kernel: Kernel logging (proc) stopped. >> Oct 13 06:25:01 srvc15 kernel: imklog 4.4.2, log source = /proc/kmsg >> started. >> Oct 13 06:25:01 srvc15 rsyslogd: [origin software="rsyslogd" >> swVersion="4.4.2" x-pid="2075" x-info="http://www.rsyslog.com";] (re)start >> >> From the container: >> Oct 10 06:25:01 srvc15c0 kernel: imklog 3.18.6, log source = /proc/kmsg >> started. >> Oct 10 06:25:01 srvc15c0 rsyslogd: [origin software="rsyslogd" >> swVersion="3.18.6" x-pid="299" x-info="http://www.rsyslog.com";] restart >> Oct 11 06:25:01 srvc15c0 kernel: imklog 3.18.6, log source = /proc/kmsg >> started. >> Oct 11 06:25:01 srvc15c0 rsyslogd: [origin software="rsyslogd" >> swVersion="3.18.6" x-pid="299" x-info="http://www.rsyslog.com";] restart >> Oct 12 06:25:01 srvc15c0 kernel: imklog 3.18.6, log source = /proc/kmsg >> started. >> Oct 12 06:25:01 srvc15c0 rsyslogd: [origin software="rsyslogd" >> swVersion="3.18.6" x-pid="299" x-info="http://www.rsyslog.com";] restart >> Oct 13 06:25:01 srvc15c0 kernel: imklog 3.18.6, log source = /proc/kmsg >> started. >> Oct 13 06:25:01 srvc15c0 rsyslogd: [origin software="rsyslogd" >> swVersion="3.18.6" x-pid="299" x-info="http://www.rsyslog.com";] restart >> >> That's all :( >> Jorge E. Espada >> >> >> >> >> On Wed, Oct 13, 2010 at 9:45 AM, Daniel Lezcano wrote: >> >>> On 10/13/2010 02:42 PM, jorge espada wrote: >>> >>>> The error was ICMP "no route to host", but now i change the conf and >>>> instead of vepa I used bridge...after destroy and create the containers >>>> with >>>> the new conf, seems to be working as expected for a while, but from time >>>> to >>>> time "the container became unresponsive".. no ping (100% packet lost). >>>> The >>>> curious thing is, if I login into the host and then to the container >>>> (lxc-console) and ping another container or host the " unresponsive >>>> container " become alive again (ping OK)so I'm a bit lost.. >>>> >>> >>> Is there any trace in /var/log/messages ? >>> >> >> > -- Download new Adobe(R) Flash(R) Builder(TM) 4 The new Adobe(R) Flex(R) 4 and Flash(R) Builder(TM) 4 (formerly Flex(R) Builder(TM)) enable the development of rich applications that run across multiple browsers and platforms. Download your free trials today! http://p.sf.net/sfu/adobe-dev2dev___ Lxc-devel mailing list Lxc-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lxc-devel
