You can subscribe to this list here.
2010 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(18) |
Dec
(7) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
2011 |
Jan
(21) |
Feb
(8) |
Mar
(8) |
Apr
(7) |
May
(1) |
Jun
(1) |
Jul
(5) |
Aug
(42) |
Sep
(1) |
Oct
|
Nov
|
Dec
|
2012 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(8) |
Sep
(13) |
Oct
|
Nov
|
Dec
|
2013 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(2) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2015 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(1) |
Oct
(1) |
Nov
|
Dec
|
From: Jon A. H. <jo...@gm...> - 2011-01-26 22:37:39
|
Hello Carachi! 2011/1/26 Carachi <car...@gm...>: > Thank you JonAn > > Tomorrow I will send you all. > > a question : Can I increase also the performace of my web server (apache + > php) with Kestrel? There are some plataforms which allow you to do so. But by now we only support MPI (and in a close future Charm++). If you want we can work together to write a Pacemaker module : https://wiki.ubuntu.com/ClusterStack/LucidTesting Regards, JonAn. |
From: Carachi <car...@gm...> - 2011-01-26 22:37:38
|
Thank you JonAn Tomorrow I will send you all. a question : Can I increase also the performace of my web server (apache + php) with Kestrel? Thank you a lot Bye On 26/01/2011 16:31, Jon Ander Hernandez wrote: > Ummm... the log is normal, but as you said the mac is empty. > It seems that there is something extrange with dnsmasq, maybe there is > something else modifying its configuration or something... > > Can you check/send me this file? > > /var/lib/misc/dnsmasq.leases > > And the configuration file: > > /etc/dnsmasq.conf > > Regards, > > JonAn. |
From: Jon A. H. <jo...@gm...> - 2011-01-26 15:31:56
|
Ummm... the log is normal, but as you said the mac is empty. It seems that there is something extrange with dnsmasq, maybe there is something else modifying its configuration or something... Can you check/send me this file? /var/lib/misc/dnsmasq.leases And the configuration file: /etc/dnsmasq.conf Regards, JonAn. |
From: Carachi <car...@gm...> - 2011-01-26 14:56:50
|
This is the log. On 26/01/2011 15:18, Jon Ander Hernandez wrote: > Hi Carachi, > > I need to fix an issue which makes empty events which is what the log > shows (somehow I send "\n" chars to the daemon's FIFO...), but they > have no effect on the system, the interesting bits should be before > that event. :-S > > Please send the whole log. > > The last error is because kestrel-daemon what stopped after > kestrel-rpc (kestrel-rpc closed the FIFO...), I will fix it so it > doesn't show the message again. > > Regards, > > JonAn. |
From: Jon A. H. <jo...@gm...> - 2011-01-26 14:18:31
|
Hi Carachi, I need to fix an issue which makes empty events which is what the log shows (somehow I send "\n" chars to the daemon's FIFO...), but they have no effect on the system, the interesting bits should be before that event. :-S Please send the whole log. The last error is because kestrel-daemon what stopped after kestrel-rpc (kestrel-rpc closed the FIFO...), I will fix it so it doesn't show the message again. Regards, JonAn. |
From: Carachi <car...@gm...> - 2011-01-26 13:42:07
|
Hi Jon thnk you for your reply. This is the register and the log file. sudo kestrel-control --register group1 Turn now on the nodes to add them to the group "image1" Press crtl+c when you are done Registering node "kestrel-group1-1" mac : ip : 192.168.30.26 Registering node "kestrel-group1-1" mac : ip : 192.168.30.8 Log file: ====================== mar 25 gen 2011, 17.27.13, CET Event's parameters : action : mac : ip : hostname : ---------------------- Unexpected event. This node is not a registered node. ====================== /usr/sbin/kestrel-daemon: riga 79: /var/lib/kestrel/rpc/fifo: Chiamata di sistema interrotta ====================== Thank you Diego On 26/01/2011 03:19, Jon Ander Hernandez wrote: > 2011/1/25 Carachi<car...@gm...>: >> Thank you Jon! >> Now I finish the installation, I created the image and now I want register >> the slave node, but it show without the mac-address and so it doesn't >> register the slave node >> >> >> sudo kestrel-control --register group1 >> >> Turn now on the nodes to add them to the group "image1" >> >> Press crtl+c when you are done >> >> Registering node "kestrel-group1-1" >> mac : >> ip : 192.168.30.26 > I can't reproduce it here, can you show me the last lines of the > kestrel log file? > > less /var/log/kestrel-daemon.log > > Regards, > > JonAn. |
From: Jon A. H. <jo...@gm...> - 2011-01-26 02:19:39
|
2011/1/25 Carachi <car...@gm...>: > Thank you Jon! > Now I finish the installation, I created the image and now I want register > the slave node, but it show without the mac-address and so it doesn't > register the slave node > > > sudo kestrel-control --register group1 > > Turn now on the nodes to add them to the group "image1" > > Press crtl+c when you are done > > Registering node "kestrel-group1-1" > mac : > ip : 192.168.30.26 I can't reproduce it here, can you show me the last lines of the kestrel log file? less /var/log/kestrel-daemon.log Regards, JonAn. |
From: Carachi <car...@gm...> - 2011-01-25 17:40:15
|
Thank you Jon! Now I finish the installation, I created the image and now I want register the slave node, but it show without the mac-address and so it doesn't register the slave node sudo kestrel-control --register group1 Turn now on the nodes to add them to the group "image1" Press crtl+c when you are done Registering node "kestrel-group1-1" mac : ip : 192.168.30.26 Thank you a lot On 25/01/2011 17:26, Jon Ander Hernandez wrote: > Install syslinux package : > > sudo apt-get install syslinux > > Regards, > > JonAn. |
From: Jon A. H. <jo...@gm...> - 2011-01-25 16:26:41
|
Install syslinux package : sudo apt-get install syslinux Regards, JonAn. |
From: Carachi <car...@gm...> - 2011-01-25 16:22:51
|
Hi Jon Ander Hernandez thank for your quickly reply. this is the post if now I try to reinstall the package. sudo apt-get install kestrelhpc kestrelhpc-openmpi kestrelhpc-ganglia Lettura elenco dei pacchetti... Fatto Generazione albero delle dipendenze Lettura informazioni sullo stato... Fatto kestrelhpc è già alla versione più recente. kestrelhpc-ganglia è già alla versione più recente. kestrelhpc-openmpi è già alla versione più recente. 0 aggiornati, 0 installati, 0 da rimuovere e 3 non aggiornati. 3 non completamente installati o rimossi. Dopo quest'operazione, verranno occupati 0B di spazio su disco. Continuare [S/n]? Configurazione di kestrelhpc (1.9.13-2011-01-20-23-38-54)... *cp: impossibile eseguire stat di "/usr/lib/syslinux/pxelinux.0": File o directory non esistente* *dpkg: errore nell'elaborare kestrelhpc (--configure): il sottoprocesso vecchio script di post-installation ha restituito lo stato di errore 1* dpkg: problemi con le dipendenze impediscono la configurazione di kestrelhpc-ganglia: kestrelhpc-ganglia dipende da kestrelhpc; comunque: Il pacchetto kestrelhpc non è ancora configurato. dpkg: errore nell'elaborare kestrelhpc-ganglia (--configure): problemi con le dipendenze - lasciato non configurato dpkg: problemi con le dipendenze impediscono la configurazione di kestrelhpc-openmpi: kestrelhpc-openmpi dipende da kestrelhpc; comunque: Il pacchetto kestrelhpc non è ancora configurato. dpkg: errore nell'elaborare kestrelhpc-openmpi (--configure): problemi con le dipendenze - lasciato non configurato Segnalazione apport non scritta poiché il messaggio di errore indica la presenza di un fallimento precedente. Segnalazione apport non scritta poiché il messaggio di errore indica la presenza di un fallimento precedente. Si sono verificati degli errori nell'elaborazione: kestrelhpc kestrelhpc-ganglia kestrelhpc-openmpi E: Sub-process /usr/bin/dpkg returned an error code (1) Have you need other? Thank you On 25/01/2011 17:14, Jon Ander Hernandez wrote: > Hi Diego, > > 2011/1/25 Carachi<car...@gm...>: >> I try to reinstall all. >> I installed Ubuntu 10.10 i386 (is egual to x64 or is better 64bit??), and >> after reboot Iaunch this command: >> >> sudo apt-get install python-software-properties >> sudo add-apt-repository ppa:kestrel/kestrelhpc >> >> sudo su >> echo "deb http://ppa.launchpad.net/kestrel/kestrelhpc/ubuntu maverick main" >>>> /etc/apt/source.list >> echo "deb-src http://ppa.launchpad.net/kestrel/kestrelhpc/ubuntu maverick >> main">> /etc/apt/source.list >> apt-get update >> apt-get install kestrelhpc kestrelhpc-openmpi kestrelhpc-ganglia >> >> but the installation fail with this error: >> >> >> /boot/dracut.img-2.6.35-22-generic-pae does not exist. Cannot update. > This is a warning, not an error. > >> Si sono verificati degli errori nell'elaborazione: >> kestrelhpc >> kestrelhpc-ganglia >> kestrelhpc-openmpi >> E: Sub-process /usr/bin/dpkg returned an error code (1) >> How can I solve it? > There must be something else, please post me the whole ouput. :-) > > Regards, > > JonAn. |
From: Jon A. H. <jo...@gm...> - 2011-01-25 16:14:28
|
Hi Diego, 2011/1/25 Carachi <car...@gm...>: > I try to reinstall all. > I installed Ubuntu 10.10 i386 (is egual to x64 or is better 64bit??), and > after reboot Iaunch this command: > > sudo apt-get install python-software-properties > sudo add-apt-repository ppa:kestrel/kestrelhpc > > sudo su > echo "deb http://ppa.launchpad.net/kestrel/kestrelhpc/ubuntu maverick main" >>> /etc/apt/source.list > echo "deb-src http://ppa.launchpad.net/kestrel/kestrelhpc/ubuntu maverick > main" >> /etc/apt/source.list > apt-get update > apt-get install kestrelhpc kestrelhpc-openmpi kestrelhpc-ganglia > > but the installation fail with this error: > > > /boot/dracut.img-2.6.35-22-generic-pae does not exist. Cannot update. This is a warning, not an error. > Si sono verificati degli errori nell'elaborazione: > kestrelhpc > kestrelhpc-ganglia > kestrelhpc-openmpi > E: Sub-process /usr/bin/dpkg returned an error code (1) > How can I solve it? There must be something else, please post me the whole ouput. :-) Regards, JonAn. |
From: Carachi <car...@gm...> - 2011-01-25 16:06:45
|
I try to reinstall all. I installed Ubuntu 10.10 i386 (is egual to x64 or is better 64bit??), and after reboot Iaunch this command: |sudo apt-get install python-software-properties sudo add-apt-repository ppa:kestrel/kestrelhpc | |sudo su echo "deb http://ppa.launchpad.net/kestrel/kestrelhpc/ubuntu maverick main" >> /etc/apt/source.list echo "deb-src http://ppa.launchpad.net/kestrel/kestrelhpc/ubuntu maverick main" >> /etc/apt/source.list apt-get update apt-get install |kestrelhpc kestrelhpc-openmpi kestrelhpc-ganglia but the installation fail with this error: /boot/dracut.img-2.6.35-22-generic-pae does not exist. Cannot update. Si sono verificati degli errori nell'elaborazione: kestrelhpc kestrelhpc-ganglia kestrelhpc-openmpi E: Sub-process /usr/bin/dpkg returned an error code (1) How can I solve it? Thank you Diego On 25/01/2011 14:45, Jon Ander Hernandez wrote: > 2011/1/25 Carachi<car...@gm...>: >> Hi Jon Ander Hernandez, >> when i launch >> >> sudo /etc/init.d/dnsmasq restart >> >> it fail. >> >> I check also the configuration file (dnsmasq --test) but there isn't >> error... >> (also the PXE boot of the node fail!) >> >> what is wrong? > Try reinstalling dnsmasq and testing again dnsmasq : > > sudo apt-get --purge --reinstall install dnsmasq > > sudo /etc/init.d/dnsmasq restart > > If it works you can then reconfigure it with kestrel : > > sudo kestrel-reconfigure --frontend > > If it doesn't work send me dnsmasq's log file : > > cat /var/log/syslog | grep dnsmasq | grep "Jan 25"> debug.log > > Regards, > > JonAn |
From: Jon A. H. <jo...@gm...> - 2011-01-25 13:45:55
|
2011/1/25 Carachi <car...@gm...>: > Hi Jon Ander Hernandez, > when i launch > > sudo /etc/init.d/dnsmasq restart > > it fail. > > I check also the configuration file (dnsmasq --test) but there isn't > error... > (also the PXE boot of the node fail!) > > what is wrong? Try reinstalling dnsmasq and testing again dnsmasq : sudo apt-get --purge --reinstall install dnsmasq sudo /etc/init.d/dnsmasq restart If it works you can then reconfigure it with kestrel : sudo kestrel-reconfigure --frontend If it doesn't work send me dnsmasq's log file : cat /var/log/syslog | grep dnsmasq | grep "Jan 25" > debug.log Regards, JonAn |
From: Carachi <car...@gm...> - 2011-01-25 11:16:14
|
Hi Jon Ander Hernandez, when i launch sudo /etc/init.d/dnsmasq restart it fail. I check also the configuration file (dnsmasq --test) but there isn't error... (also the PXE boot of the node fail!) what is wrong? Thank you On 25/01/2011 04:02, Jon Ander Hernandez wrote: > Hi Carachi! > > 2011/1/24 Carachi<car...@gm...>: >> Hi, >> I try to install kestrelHPC but I have a problem with dnsmasq. >> I follow the documentation and the tftp server don't start. >> Why? Can you help me? > Is dnsmasq starting correctly? > > sudo /etc/init.d/dnsmasq restart > > Or do the nodes fail during the PXE boot? > > Regards, > > JonAn. |
From: Jon A. H. <jo...@gm...> - 2011-01-25 03:02:33
|
Hi Carachi! 2011/1/24 Carachi <car...@gm...>: > Hi, > I try to install kestrelHPC but I have a problem with dnsmasq. > I follow the documentation and the tftp server don't start. > Why? Can you help me? Is dnsmasq starting correctly? sudo /etc/init.d/dnsmasq restart Or do the nodes fail during the PXE boot? Regards, JonAn. |
From: Carachi <car...@gm...> - 2011-01-24 17:52:46
|
Hi, I try to install kestrelHPC but I have a problem with dnsmasq. I follow the documentation and the tftp server don't start. Why? Can you help me? Thank you Bye |
From: Dipietro S. <dip...@gm...> - 2011-01-24 17:51:52
|
Hi, I try to install kestrelHPC but I have a problem with dnsmasq. I follow the documentation and the tftp server don't start. Why? Can you help me? Thank you Bye |
From: Denis S. A. <dar...@gm...> - 2011-01-13 14:35:31
|
Using the command sudo kestrel-users --add-to-cluster bob where bob is a user in group users results in output like: Adding user `bob' to group `kestrel' ... Adding user bob to group kestrel Done. chown: invalid group: `bob:bob' Any idea what's going on there? I have checked the kestrel-users script and with the option -add-to-cluster, the user, bob in this case, is created with the group kestrel, but that means that there are two bob users in the system, 1 with the group bob, in ubuntu, and the other one with kestrel. Wouldn't it better to just change that users group?? |
From: Denis S. A. <dar...@gm...> - 2011-01-10 16:04:23
|
Hello there, I want to give a push to the mailing list, so there it goes. There is "bug" in the web page, in the documentation section -> Man pages, every time you click on a command to see its man page, the title of the web, not in the explorer window but the image of KestrelHPC at the top of the web, disappears. Cheers, Denis |
From: Jon A. H. <jo...@gm...> - 2010-12-08 21:31:16
|
2010/12/8 Jon Ander Hernandez <jo...@gm...>: > On Saturday I fixed our Dracut packages. > > https://launchpad.net/~kestrel/+archive/kestrelhpc/+packages I forgot to mention that now we have nfs4 working perfectly :-) |
From: Jon A. H. <jo...@gm...> - 2010-12-08 21:28:00
|
2010/12/8 Denis Sánchez Argoitia <dar...@gm...>: > This weekend I will check all the commands that you mentioned and the fifo. > Jairo may try the kestrel-control command in the cluster in the following > days. > > about the dhcp I might have an idea to solve the problem in an easier way, Please explain it in more detail. > but first I need to check deeply the commands. I will have you informed. Here is a more in depth explanation of how it works, and the problems we address : * If a node starts and we haven't started the node explicitly, it must not start as a node of the cluster. Maybe the node is a computer of a computer room which acts at nights as a cluster and a user wants to use it as a common computer, so the pxe config should be configured for it. * For registering a node as a Group of nodes, is as simple as creating a link which points the mac of the node in the pxe dir to the group. This is the status of a default kestrel tftpboot dir : $ ls -l /var/lib/kestrel/tftpboot{*,*/*} vmlinuz dracut-img initrd.img # kernel and initramfs files pxelinux.0 vesamenu.c32 reboot.c32 splash.png # pxefiles pxelinux.cfg/default -> ../reboot # the default machine load the reboot config by default # the machine with these macs load the config group1 pxelinux.cfg/01-00-0c-6e-64-d8-b -> ../group1 pxelinux.cfg/01-00-0c-6e-64-d8-c -> ../group1 pxelinux.cfg/01-00-0c-6e-64-d8-d -> ../group1 pxelinux.cfg/01-00-0c-6e-64-d8-e -> ../group2 pxelinux.cfg/01-00-0c-6e-64-d8-f -> ../group2 group1 -> reboot # node image's config files : (the .register file is the same as the normal one, except that the cmdline includes the register parameter) test1 test1.register test2 test2.register * So when you start a group of nodes, you should the group1 link point to test1, and afterwards send the wol signal to those nodes. * When you register a new computer, you make pxelinux.cfg/default point to "image name".register So the script "node/register.d/tftpboot" makes a node point to the group when we receive a register event (which is executed by kestrel-daemon) : http://kestrelhpc.svn.sourceforge.net/viewvc/kestrelhpc/Kestrelhpc/Src/share/node/register.d/tftpboot One of the hard things to do with dhcp, is that when you register a new computer, you must use somekind of barrier since kestrel-daemon run the register.d scripts in parallel. Take a look to the kestrel-daemons source (line 57) : http://kestrelhpc.svn.sourceforge.net/viewvc/kestrelhpc/Kestrelhpc/Src/kestrel-daemon?revision=78&view=markup If you have any doubt pls ask 0:-) JonAn. |
From: Denis S. A. <dar...@gm...> - 2010-12-08 16:22:40
|
This weekend I will check all the commands that you mentioned and the fifo. Jairo may try the kestrel-control command in the cluster in the following days. about the dhcp I might have an idea to solve the problem in an easier way, but first I need to check deeply the commands. I will have you informed. Cheers, Denis 2010/12/8 Jon Ander Hernandez <jo...@gm...> > On Saturday I fixed our Dracut packages. > > https://launchpad.net/~kestrel/+archive/kestrelhpc/+packages<https://launchpad.net/%7Ekestrel/+archive/kestrelhpc/+packages> > > > I also added some documentation about the node install and configure > scripts : > > - > http://kestrelhpc.svn.sourceforge.net/viewvc/kestrelhpc/Kestrelhpc/Src/share/node/install.d/README > > - > http://kestrelhpc.svn.sourceforge.net/viewvc/kestrelhpc/Kestrelhpc/Src/share/node/configure.d/README > > > We need to finish kestrel-control and kestrel-daemon : > > - kestrel-control needs to write a wake event the fifo file > /var/lib/kestrel/rpc/fifo before trying to wake up clients. > > - check kestrel-control --wake|--shutdown|--reboot works on a real cluster. > > - kestrel-daemon and register.d scripts need to be fixed, to register > the nodes when we are registering them with kestrel-control > --register. > > I actually have a design problem, because I use the dhcp config file > to link the mac of a node to the node image os. But this limits us to > use a node only with only a node image os. So instead we should create > cluster node groups with kestrel-control --register [node-group] and > link a node group to 1 or more node image oses. > > That means that we need to rewrite these functions on kestrel-lib : > > > http://kestrelhpc.svn.sourceforge.net/viewvc/kestrelhpc/Kestrelhpc/Src/share/kestrel-lib > > connected_nodes() > connected_images() > disconnected_nodes() > disconnected_images() > registered_nodes() > registered_images() > > > And finally we need to write kestrel-launch-mpi which should be trivial. > > > Regards, > > Jon Ander. > > > ------------------------------------------------------------------------------ > What happens now with your Lotus Notes apps - do you make another costly > upgrade, or settle for being marooned without product support? Time to move > off Lotus Notes and onto the cloud with Force.com, apps are easier to > build, > use, and manage than apps on traditional platforms. Sign up for the Lotus > Notes Migration Kit to learn more. http://p.sf.net/sfu/salesforce-d2d > _______________________________________________ > Kestrelhpc-developers mailing list > Kes...@li... > https://lists.sourceforge.net/lists/listinfo/kestrelhpc-developers > |
From: Jon A. H. <jo...@gm...> - 2010-12-08 15:22:40
|
On Saturday I fixed our Dracut packages. https://launchpad.net/~kestrel/+archive/kestrelhpc/+packages I also added some documentation about the node install and configure scripts : - http://kestrelhpc.svn.sourceforge.net/viewvc/kestrelhpc/Kestrelhpc/Src/share/node/install.d/README - http://kestrelhpc.svn.sourceforge.net/viewvc/kestrelhpc/Kestrelhpc/Src/share/node/configure.d/README We need to finish kestrel-control and kestrel-daemon : - kestrel-control needs to write a wake event the fifo file /var/lib/kestrel/rpc/fifo before trying to wake up clients. - check kestrel-control --wake|--shutdown|--reboot works on a real cluster. - kestrel-daemon and register.d scripts need to be fixed, to register the nodes when we are registering them with kestrel-control --register. I actually have a design problem, because I use the dhcp config file to link the mac of a node to the node image os. But this limits us to use a node only with only a node image os. So instead we should create cluster node groups with kestrel-control --register [node-group] and link a node group to 1 or more node image oses. That means that we need to rewrite these functions on kestrel-lib : http://kestrelhpc.svn.sourceforge.net/viewvc/kestrelhpc/Kestrelhpc/Src/share/kestrel-lib connected_nodes() connected_images() disconnected_nodes() disconnected_images() registered_nodes() registered_images() And finally we need to write kestrel-launch-mpi which should be trivial. Regards, Jon Ander. |
From: Jon A. H. <jo...@gm...> - 2010-12-08 15:03:32
|
Hi everybody : KestrelHPC is a set of tools which help setting up the basic services required for creating a simple cluster : - Create the node os image. - Set up the nfs root export. - Set up the nfs home export. - Create ssh keys for login into the nodes without passwords. - Set up a dhcp server, and a tftp server for PXE. - etc... We also wanted to be capable of reconfigure everything easily. So the scripts recreate the configuration for each service reading some variables from a config file. Those scripts are called the configuration scripts : /usr/share/kestrel/frontend/configure.d/ /usr/share/kestrel/node/configure.d/ So when we change the config file, we only need to run kestrel-reconfigure --all, and everything will be reconfigured. We also may need to do some actions when we create a new node image or when we remove it. We also have to do some actions when we install or remove kestrel. Does actions are done by the installation scripts : /usr/share/kestrel/frontend/install.d/ /usr/share/kestrel/node/install.d/ Also note, that some scripts need to be run inside a chroot under the node image. We use the suffix "_chroot" to denote that. We also have 2 daemons : kestrel-daemon and kestrel-rpc - kestrel-daemon reads events/commands from fifo file : /var/lib/kestrel/rpc/fifo. We have connect, disconnect, register, desregister, wake and shutdown events/commands : * connect and disconnect events are sent by kestrel-rpc when a client makes a rpc call to the server indicating that is has started up or is shuting down. * register event is like connect. The only diference is that the client should be registered (we are registering new nodes with kestrel-control --register). * desregister is called from kestrel-control. ToDo. * wake, shutdown are called from kestrel-control. ToDo. We need these events to do the necesary steps before starting the nodes, for example for granting access to the nodes to the nfs services through the firewall. And for the counterpart when shutting down a node. - kestrel-rpc. Is a simple rpc and easily extendable written with python. Well this was a quite informal, and quite incomplete explanation, please ask if you have any doubt. :-) Regards, Jon Ander. |
From: Denis S. A. <dar...@gm...> - 2010-12-08 14:41:25
|
I think that it is a really good explanation of why the decision of using dracut instead of the traditional init. On the other hand, dracut is more "modular" and clear, which enables us to add support for other FS such as GFS and many others. 2010/12/8 Jon Ander Hernandez <jo...@gm...> > Dracut is an udev based initramfs generator. We use it instead of a > more tradicional script using initramfs-tools, like pelicanhpc does, > because : > > - initramfs-tools doesn't support nfs4. We thought nfs4 support was > something quite important, because its cache infraestructure could be > a real gain. > > - initramfs-tool boot scripts can be are really complex. Casper is the > initramfs script used in Ubuntu live cds, which creates the writable > overlay. It also allows to boot from a nfs export, and use a > persistent overlay. > Casper main script has 722 lines : > > $ wc -l /usr/share/initramfs-tools/scripts/casper > 722 /usr/share/initramfs-tools/scripts/casper > > It also set ups lots of things during the initramfs stage : > > 14locales 18hostname 19keyboard 20xconfig 31disable_update_notifier > 32disable_hibernation 33enable_apport_crashes 34disable_kde_services > 35fix_language_selector ... > > We don't need so much things in a cluster node. > > > A dracut based approach was much simpler : > > > http://kestrelhpc.svn.sourceforge.net/viewvc/kestrelhpc/Kestrelhpc/Src/dracut/union_mount_with_aufs.diff?revision=25&content-type=text/plain&pathrev=40 > > We only needed to create a module which provides a writable overlay using > aufs. > > The only problem is that Debian Dracut packages needed to be updated, > so we did : > > http://www.spinics.net/lists/linux-initramfs/msg01625.html > > > Regards, > > Jon Ander. > > > ------------------------------------------------------------------------------ > What happens now with your Lotus Notes apps - do you make another costly > upgrade, or settle for being marooned without product support? Time to move > off Lotus Notes and onto the cloud with Force.com, apps are easier to > build, > use, and manage than apps on traditional platforms. Sign up for the Lotus > Notes Migration Kit to learn more. http://p.sf.net/sfu/salesforce-d2d > _______________________________________________ > Kestrelhpc-developers mailing list > Kes...@li... > https://lists.sourceforge.net/lists/listinfo/kestrelhpc-developers > |