From nathan.baca at gmail.com Tue Jun 2 17:49:09 2009 From: nathan.baca at gmail.com (Nathan Baca) Date: Tue, 2 Jun 2009 18:49:09 -0600 Subject: [cAos] Live install on cciss hardware Message-ID: Hello, I have been working trying to get the Caos installers (live and regular) to work with cciss hardware. I have been using the caos-nsa-base-1.0.15.x86_64.iso. Attached are a couple of patches to the Disk and Boot perl modules of the live installer. I also included the kickstart file I have been using. I think the changes to Disk.pm are pretty minor. But I did have to work around an issue in Boot.pm with grub-install silently failing. I was seeing with debug turned on: # /usr/sbin/chroot /newroot /sbin/grub-install /dev/cciss/c0d0 > expr: non-numeric argument ... OK Essentailly I modified the grub-install call to use the 'hd0' alias already created in device.map which allowed grub to install to the boot partition and allow the node to boot off the localdisk. Anybody with more grub expertise care to comment? Thanks, Nate -- Nathan Baca nathan.baca at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.infiscale.org/pipermail/caos/attachments/20090602/16c33134/attachment.html -------------- next part -------------- A non-text attachment was scrubbed... Name: ccisspatch.tar.bz2 Type: application/x-bzip2 Size: 3522 bytes Desc: not available Url : http://lists.infiscale.org/pipermail/caos/attachments/20090602/16c33134/attachment.bz2 From gmkurtzer at gmail.com Tue Jun 2 21:17:08 2009 From: gmkurtzer at gmail.com (Greg Kurtzer) Date: Tue, 2 Jun 2009 21:17:08 -0700 Subject: [cAos] Live install on cciss hardware In-Reply-To: References: Message-ID: <571f1a060906022117m267bef90jdf19719f5c23d860@mail.gmail.com> Hi Nathan, Thanks for the patch. It has been integrated and is in the nsa-testing package repository as part of the caos-utils update. Please test and make sure everything got integrated properly. Greg On Tue, Jun 2, 2009 at 5:49 PM, Nathan Baca wrote: > Hello, > > I have been working trying to get the Caos installers (live and regular) to > work with cciss hardware. I have been using the > caos-nsa-base-1.0.15.x86_64.iso. Attached are a couple of patches to the > Disk and Boot perl modules of the live installer. I also included the > kickstart file I have been using. > > I think the changes to Disk.pm are pretty minor. But I did have to work > around an issue in Boot.pm with grub-install silently failing. I was seeing > with debug turned on: > > # /usr/sbin/chroot /newroot /sbin/grub-install /dev/cciss/c0d0 >> expr: non-numeric argument > ... > ?? OK > > Essentailly I modified the grub-install call to use the 'hd0' alias already > created in device.map which allowed grub to install to the boot partition > and allow the node to boot off the localdisk. Anybody with more grub > expertise care to comment? > > Thanks, Nate > -- > Nathan Baca > nathan.baca at gmail.com > > _______________________________________________ > cAos mailing list > cAos at caoslinux.org > http://lists.caosity.org/mailman/listinfo/caos > > -- Greg Kurtzer http://www.infiscale.com/ http://www.perceus.org/ http://www.caoslinux.org/ From Olivier.Marsden at ec-lyon.fr Mon Jun 8 05:28:28 2009 From: Olivier.Marsden at ec-lyon.fr (Olivier Marsden) Date: Mon, 08 Jun 2009 14:28:28 +0200 Subject: [cAos] problem booting caos/perceus nodes Message-ID: <4A2D03EC.6090805@ec-lyon.fr> Dear Caos users, I am trying to install a small test cluster using caos+perceus (both latest versions), and am having problems provisioning the nodes. Master node installation went fine, and everything seems to work well (except for the keymap for my french keyboard, on which neither shift keys nor the alt-gr key work, but that's another question!). Sidekick configured perceus for me, if I understood correctly. When I boot a node, it correctly gets to the perceus boot menu, but then fails to provision from the master (specifically, fails to download the vnfs), with the following error: mount : 10.254.0.1:/var/lib/perceus failed, reason given by server: Permission denied The /etc/exports on the master node seems to be correct, and nfsd is running (see end of message), and the nic MAC address of my node appears in /etc/perceus/ethers . Has anyone any idea of what I might be doing incorrectly? Thanks in advance! Olivier Marsden >> cat /etc/exports /var/lib/perceus 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus export /home 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export /usr/local 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus export /usr/cports 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus export /opt 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus export /srv 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export /private 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >> /usr/sbin/rpcinfo -p program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 100011 1 udp 864 rquotad 100011 2 udp 864 rquotad 100011 1 tcp 867 rquotad 100011 2 tcp 867 rquotad 100003 2 udp 2049 nfs 100003 3 udp 2049 nfs 100003 4 udp 2049 nfs 100021 1 udp 45615 nlockmgr 100021 3 udp 45615 nlockmgr 100021 4 udp 45615 nlockmgr 100021 1 tcp 48041 nlockmgr 100021 3 tcp 48041 nlockmgr 100021 4 tcp 48041 nlockmgr 100003 2 tcp 2049 nfs 100003 3 tcp 2049 nfs 100003 4 tcp 2049 nfs 100005 1 udp 59768 mountd 100005 1 tcp 55015 mountd 100005 2 udp 59768 mountd 100005 2 tcp 55015 mountd 100005 3 udp 59768 mountd 100005 3 tcp 55015 mountd 100024 1 udp 57172 status 100024 1 tcp 59417 status From kirkby at etsu.edu Mon Jun 8 06:09:42 2009 From: kirkby at etsu.edu (Scott Kirkby) Date: Mon, 08 Jun 2009 09:09:42 -0400 Subject: [cAos] problem booting caos/perceus nodes In-Reply-To: <4A2D03EC.6090805@ec-lyon.fr> References: <4A2D03EC.6090805@ec-lyon.fr> Message-ID: <4A2D0D96.3000606@etsu.edu> Oliver, Do you have a firewall on the server interface? That may be block the nfs mount. Scott Olivier Marsden wrote: > Dear Caos users, > > I am trying to install a small test cluster using caos+perceus (both > latest versions), and > am having problems provisioning the nodes. > Master node installation went fine, and everything seems to work well > (except for the keymap for my french keyboard, on which neither shift keys > nor the alt-gr key work, but that's another question!). > Sidekick configured perceus for me, if I understood correctly. > When I boot a node, it correctly gets to the perceus boot menu, but > then fails to provision from the master (specifically, fails to > download the vnfs), > with the following error: > mount : 10.254.0.1:/var/lib/perceus failed, reason given by server: > Permission denied > > The /etc/exports on the master node seems to be correct, and nfsd is > running (see end > of message), and the nic MAC address of my node appears in > /etc/perceus/ethers . > Has anyone any idea of what I might be doing incorrectly? > Thanks in advance! > > Olivier Marsden > > > > > > > > > >> cat /etc/exports > /var/lib/perceus 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # > perceus export > /home 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export > /usr/local 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus > export > /usr/cports 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus > export > /opt 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus export > /srv 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export > /private 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export > > > >> /usr/sbin/rpcinfo -p > program vers proto port > 100000 2 tcp 111 portmapper > 100000 2 udp 111 portmapper > 100011 1 udp 864 rquotad > 100011 2 udp 864 rquotad > 100011 1 tcp 867 rquotad > 100011 2 tcp 867 rquotad > 100003 2 udp 2049 nfs > 100003 3 udp 2049 nfs > 100003 4 udp 2049 nfs > 100021 1 udp 45615 nlockmgr > 100021 3 udp 45615 nlockmgr > 100021 4 udp 45615 nlockmgr > 100021 1 tcp 48041 nlockmgr > 100021 3 tcp 48041 nlockmgr > 100021 4 tcp 48041 nlockmgr > 100003 2 tcp 2049 nfs > 100003 3 tcp 2049 nfs > 100003 4 tcp 2049 nfs > 100005 1 udp 59768 mountd > 100005 1 tcp 55015 mountd > 100005 2 udp 59768 mountd > 100005 2 tcp 55015 mountd > 100005 3 udp 59768 mountd > 100005 3 tcp 55015 mountd > 100024 1 udp 57172 status > 100024 1 tcp 59417 status > > _______________________________________________ > cAos mailing list > cAos at caoslinux.org > http://lists.caosity.org/mailman/listinfo/caos > From Olivier.Marsden at ec-lyon.fr Mon Jun 8 08:05:40 2009 From: Olivier.Marsden at ec-lyon.fr (Olivier Marsden) Date: Mon, 08 Jun 2009 17:05:40 +0200 Subject: [cAos] problem booting caos/perceus nodes In-Reply-To: <4A2D0D96.3000606@etsu.edu> References: <4A2D03EC.6090805@ec-lyon.fr> <4A2D0D96.3000606@etsu.edu> Message-ID: <4A2D28C4.3000403@ec-lyon.fr> I'm not sure exactly what caos uses in the firewall domain, but I was supposing that it would at least not interfere with perceus? I'll look into it, thanks for the suggestion. Olivier Scott Kirkby wrote: > Oliver, > > Do you have a firewall on the server interface? That may be block the > nfs mount. > > Scott > > Olivier Marsden wrote: > >> Dear Caos users, >> >> I am trying to install a small test cluster using caos+perceus (both >> latest versions), and >> am having problems provisioning the nodes. >> Master node installation went fine, and everything seems to work well >> (except for the keymap for my french keyboard, on which neither shift keys >> nor the alt-gr key work, but that's another question!). >> Sidekick configured perceus for me, if I understood correctly. >> When I boot a node, it correctly gets to the perceus boot menu, but >> then fails to provision from the master (specifically, fails to >> download the vnfs), >> with the following error: >> mount : 10.254.0.1:/var/lib/perceus failed, reason given by server: >> Permission denied >> >> The /etc/exports on the master node seems to be correct, and nfsd is >> running (see end >> of message), and the nic MAC address of my node appears in >> /etc/perceus/ethers . >> Has anyone any idea of what I might be doing incorrectly? >> Thanks in advance! >> >> Olivier Marsden >> >> >> >> >> >> >> >> >> >> cat /etc/exports >> /var/lib/perceus 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # >> perceus export >> /home 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >> /usr/local 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus >> export >> /usr/cports 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus >> export >> /opt 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus export >> /srv 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >> /private 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >> >> >> >> /usr/sbin/rpcinfo -p >> program vers proto port >> 100000 2 tcp 111 portmapper >> 100000 2 udp 111 portmapper >> 100011 1 udp 864 rquotad >> 100011 2 udp 864 rquotad >> 100011 1 tcp 867 rquotad >> 100011 2 tcp 867 rquotad >> 100003 2 udp 2049 nfs >> 100003 3 udp 2049 nfs >> 100003 4 udp 2049 nfs >> 100021 1 udp 45615 nlockmgr >> 100021 3 udp 45615 nlockmgr >> 100021 4 udp 45615 nlockmgr >> 100021 1 tcp 48041 nlockmgr >> 100021 3 tcp 48041 nlockmgr >> 100021 4 tcp 48041 nlockmgr >> 100003 2 tcp 2049 nfs >> 100003 3 tcp 2049 nfs >> 100003 4 tcp 2049 nfs >> 100005 1 udp 59768 mountd >> 100005 1 tcp 55015 mountd >> 100005 2 udp 59768 mountd >> 100005 2 tcp 55015 mountd >> 100005 3 udp 59768 mountd >> 100005 3 tcp 55015 mountd >> 100024 1 udp 57172 status >> 100024 1 tcp 59417 status >> >> _______________________________________________ >> cAos mailing list >> cAos at caoslinux.org >> http://lists.caosity.org/mailman/listinfo/caos >> >> > > _______________________________________________ > cAos mailing list > cAos at caoslinux.org > http://lists.caosity.org/mailman/listinfo/caos > > From gmkurtzer at gmail.com Mon Jun 8 10:19:39 2009 From: gmkurtzer at gmail.com (Greg Kurtzer) Date: Mon, 8 Jun 2009 10:19:39 -0700 Subject: [cAos] problem booting caos/perceus nodes In-Reply-To: <4A2D28C4.3000403@ec-lyon.fr> References: <4A2D03EC.6090805@ec-lyon.fr> <4A2D0D96.3000606@etsu.edu> <4A2D28C4.3000403@ec-lyon.fr> Message-ID: <571f1a060906081019x27af0ba6uf6c83e40541fd74c@mail.gmail.com> Check to make sure that NFS started up properly # /etc/init.d/nfs status # /etc/init.d/nfs restart Also, check out the iptables ruleset # iptables -L and stop IPTables if necessary for testing: # /etc/init.d/iptables stop # /sbin/chkconfig iptables off Is the keyboard having issues in the text based virtual terminals, or the graphical environment? Good luck! On Mon, Jun 8, 2009 at 8:05 AM, Olivier Marsden wrote: > I'm not sure exactly what caos uses in the firewall domain, but I was > supposing that it would at least not interfere with perceus? > I'll look into it, thanks for the suggestion. > Olivier > > Scott Kirkby wrote: >> Oliver, >> >> Do you have a firewall on the server interface? That may be block the >> nfs mount. >> >> Scott >> >> Olivier Marsden wrote: >> >>> Dear Caos users, >>> >>> I am trying to install a small test cluster using caos+perceus (both >>> latest versions), and >>> am having problems provisioning the nodes. >>> Master node installation went fine, and everything seems to work well >>> (except for the keymap for my french keyboard, on which neither shift keys >>> ?nor the alt-gr key work, but that's another question!). >>> Sidekick configured perceus for me, if I understood correctly. >>> When I boot a node, it correctly gets to the perceus boot menu, but >>> then ?fails to provision from the master (specifically, fails to >>> download the vnfs), >>> with the following error: >>> mount : 10.254.0.1:/var/lib/perceus failed, reason given by server: >>> Permission denied >>> >>> The /etc/exports on the master node seems to be correct, and nfsd is >>> running (see end >>> of message), and the nic MAC address of my node appears in >>> /etc/perceus/ethers . >>> Has anyone any idea of what I might be doing incorrectly? >>> Thanks in advance! >>> >>> Olivier Marsden >>> >>> >>> >>> >>> >>> >>> >>> >>> ?>> cat /etc/exports >>> /var/lib/perceus 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # >>> perceus export >>> /home 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>> /usr/local 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus >>> export >>> /usr/cports 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus >>> export >>> /opt 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus export >>> /srv 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>> /private 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>> >>> >>> ?>> /usr/sbin/rpcinfo -p >>> ? ?program vers proto ? port >>> ? ? 100000 ? ?2 ? tcp ? ?111 ?portmapper >>> ? ? 100000 ? ?2 ? udp ? ?111 ?portmapper >>> ? ? 100011 ? ?1 ? udp ? ?864 ?rquotad >>> ? ? 100011 ? ?2 ? udp ? ?864 ?rquotad >>> ? ? 100011 ? ?1 ? tcp ? ?867 ?rquotad >>> ? ? 100011 ? ?2 ? tcp ? ?867 ?rquotad >>> ? ? 100003 ? ?2 ? udp ? 2049 ?nfs >>> ? ? 100003 ? ?3 ? udp ? 2049 ?nfs >>> ? ? 100003 ? ?4 ? udp ? 2049 ?nfs >>> ? ? 100021 ? ?1 ? udp ?45615 ?nlockmgr >>> ? ? 100021 ? ?3 ? udp ?45615 ?nlockmgr >>> ? ? 100021 ? ?4 ? udp ?45615 ?nlockmgr >>> ? ? 100021 ? ?1 ? tcp ?48041 ?nlockmgr >>> ? ? 100021 ? ?3 ? tcp ?48041 ?nlockmgr >>> ? ? 100021 ? ?4 ? tcp ?48041 ?nlockmgr >>> ? ? 100003 ? ?2 ? tcp ? 2049 ?nfs >>> ? ? 100003 ? ?3 ? tcp ? 2049 ?nfs >>> ? ? 100003 ? ?4 ? tcp ? 2049 ?nfs >>> ? ? 100005 ? ?1 ? udp ?59768 ?mountd >>> ? ? 100005 ? ?1 ? tcp ?55015 ?mountd >>> ? ? 100005 ? ?2 ? udp ?59768 ?mountd >>> ? ? 100005 ? ?2 ? tcp ?55015 ?mountd >>> ? ? 100005 ? ?3 ? udp ?59768 ?mountd >>> ? ? 100005 ? ?3 ? tcp ?55015 ?mountd >>> ? ? 100024 ? ?1 ? udp ?57172 ?status >>> ? ? 100024 ? ?1 ? tcp ?59417 ?status >>> >>> _______________________________________________ >>> cAos mailing list >>> cAos at caoslinux.org >>> http://lists.caosity.org/mailman/listinfo/caos >>> >>> >> >> _______________________________________________ >> cAos mailing list >> cAos at caoslinux.org >> http://lists.caosity.org/mailman/listinfo/caos >> >> > > _______________________________________________ > cAos mailing list > cAos at caoslinux.org > http://lists.caosity.org/mailman/listinfo/caos > -- Greg Kurtzer http://www.infiscale.com/ http://www.perceus.org/ http://www.caoslinux.org/ From Olivier.Marsden at ec-lyon.fr Mon Jun 15 23:02:25 2009 From: Olivier.Marsden at ec-lyon.fr (Olivier Marsden) Date: Tue, 16 Jun 2009 08:02:25 +0200 Subject: [cAos] problem booting caos/perceus nodes In-Reply-To: <571f1a060906081019x27af0ba6uf6c83e40541fd74c@mail.gmail.com> References: <4A2D03EC.6090805@ec-lyon.fr> <4A2D0D96.3000606@etsu.edu> <4A2D28C4.3000403@ec-lyon.fr> <571f1a060906081019x27af0ba6uf6c83e40541fd74c@mail.gmail.com> Message-ID: <4A373571.3070308@ec-lyon.fr> Right, after a few days, I got some time to look into it again. And like any self-respecting researcher, I decided that changing only one parameter at a time was boring :-) So, with an updated system (caos updated via smart), I now get different problems, surprise surprise! I managed to boot up the nodes once, apparently correctly, and now they get stuck at the tftp phase with a tftp open timeout. The nodes receive an ip address from the dchp server, in fact the address that I specified in the ipaddr module. In my messages log on the server, I get the following lines: Jun 16 07:45:33 selene perceus-dnsmasq[2854]: DHCPDISCOVER(eth0) 00:1f:29:ea:e1:ba Jun 16 07:45:33 selene perceus-dnsmasq[2854]: DHCPOFFER(eth0) 10.254.0.12 00:1f:29:ea:e1:ba Jun 16 07:45:37 selene perceus-dnsmasq[2854]: DHCPREQUEST(eth0) 10.254.0.12 00:1f:29:ea:e1:ba Jun 16 07:45:37 selene perceus-dnsmasq[2854]: DHCPACK(eth0) 10.254.0.12 00:1f:29:ea:e1:ba node02 Any ideas? Thanks, Olivier PS: the keyboard problems are both in the xfce gui & from the command line, but all is peachy if I ssh in from another machine. Greg Kurtzer wrote: > Check to make sure that NFS started up properly > > # /etc/init.d/nfs status > # /etc/init.d/nfs restart > > Also, check out the iptables ruleset > > # iptables -L > > and stop IPTables if necessary for testing: > > # /etc/init.d/iptables stop > # /sbin/chkconfig iptables off > > Is the keyboard having issues in the text based virtual terminals, or > the graphical environment? > > Good luck! > > > On Mon, Jun 8, 2009 at 8:05 AM, Olivier > Marsden wrote: > >> I'm not sure exactly what caos uses in the firewall domain, but I was >> supposing that it would at least not interfere with perceus? >> I'll look into it, thanks for the suggestion. >> Olivier >> >> Scott Kirkby wrote: >> >>> Oliver, >>> >>> Do you have a firewall on the server interface? That may be block the >>> nfs mount. >>> >>> Scott >>> >>> Olivier Marsden wrote: >>> >>> >>>> Dear Caos users, >>>> >>>> I am trying to install a small test cluster using caos+perceus (both >>>> latest versions), and >>>> am having problems provisioning the nodes. >>>> Master node installation went fine, and everything seems to work well >>>> (except for the keymap for my french keyboard, on which neither shift keys >>>> nor the alt-gr key work, but that's another question!). >>>> Sidekick configured perceus for me, if I understood correctly. >>>> When I boot a node, it correctly gets to the perceus boot menu, but >>>> then fails to provision from the master (specifically, fails to >>>> download the vnfs), >>>> with the following error: >>>> mount : 10.254.0.1:/var/lib/perceus failed, reason given by server: >>>> Permission denied >>>> >>>> The /etc/exports on the master node seems to be correct, and nfsd is >>>> running (see end >>>> of message), and the nic MAC address of my node appears in >>>> /etc/perceus/ethers . >>>> Has anyone any idea of what I might be doing incorrectly? >>>> Thanks in advance! >>>> >>>> Olivier Marsden >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >> cat /etc/exports >>>> /var/lib/perceus 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # >>>> perceus export >>>> /home 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>>> /usr/local 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus >>>> export >>>> /usr/cports 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus >>>> export >>>> /opt 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus export >>>> /srv 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>>> /private 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>>> >>>> >>>> >> /usr/sbin/rpcinfo -p >>>> program vers proto port >>>> 100000 2 tcp 111 portmapper >>>> 100000 2 udp 111 portmapper >>>> 100011 1 udp 864 rquotad >>>> 100011 2 udp 864 rquotad >>>> 100011 1 tcp 867 rquotad >>>> 100011 2 tcp 867 rquotad >>>> 100003 2 udp 2049 nfs >>>> 100003 3 udp 2049 nfs >>>> 100003 4 udp 2049 nfs >>>> 100021 1 udp 45615 nlockmgr >>>> 100021 3 udp 45615 nlockmgr >>>> 100021 4 udp 45615 nlockmgr >>>> 100021 1 tcp 48041 nlockmgr >>>> 100021 3 tcp 48041 nlockmgr >>>> 100021 4 tcp 48041 nlockmgr >>>> 100003 2 tcp 2049 nfs >>>> 100003 3 tcp 2049 nfs >>>> 100003 4 tcp 2049 nfs >>>> 100005 1 udp 59768 mountd >>>> 100005 1 tcp 55015 mountd >>>> 100005 2 udp 59768 mountd >>>> 100005 2 tcp 55015 mountd >>>> 100005 3 udp 59768 mountd >>>> 100005 3 tcp 55015 mountd >>>> 100024 1 udp 57172 status >>>> 100024 1 tcp 59417 status >>>> >>>> _______________________________________________ >>>> cAos mailing list >>>> cAos at caoslinux.org >>>> http://lists.caosity.org/mailman/listinfo/caos >>>> >>>> >>>> >>> _______________________________________________ >>> cAos mailing list >>> cAos at caoslinux.org >>> http://lists.caosity.org/mailman/listinfo/caos >>> >>> >>> >> _______________________________________________ >> cAos mailing list >> cAos at caoslinux.org >> http://lists.caosity.org/mailman/listinfo/caos >> >> > > > > From Olivier.Marsden at ec-lyon.fr Fri Jun 19 04:06:37 2009 From: Olivier.Marsden at ec-lyon.fr (Olivier Marsden) Date: Fri, 19 Jun 2009 13:06:37 +0200 Subject: [cAos] problem booting caos/perceus nodes In-Reply-To: <4A373571.3070308@ec-lyon.fr> References: <4A2D03EC.6090805@ec-lyon.fr> <4A2D0D96.3000606@etsu.edu> <4A2D28C4.3000403@ec-lyon.fr> <571f1a060906081019x27af0ba6uf6c83e40541fd74c@mail.gmail.com> <4A373571.3070308@ec-lyon.fr> Message-ID: <4A3B713D.1090305@ec-lyon.fr> Hello again, and sorry to mug people with my problems! I have tried to test the tftp server by booting a node on harddisk, configuring it with a dhcp-obtained ip , and downloading pxelinux.0 by hand with tftp from the master : it works. This seems to indicate that the problem comes from the set-up of the nic card on the node: do people agree with this? Thanks, Olivier Olivier Marsden wrote: > Right, after a few days, I got some time to look into it again. > And like any self-respecting researcher, I decided that changing > only one parameter at a time was boring :-) > So, with an updated system (caos updated via smart), I now get > different problems, surprise surprise! I managed to boot up the > nodes once, apparently correctly, and now they get stuck at the > tftp phase with a tftp open timeout. The nodes receive an ip > address from the dchp server, in fact the address that I specified > in the ipaddr module. In my messages log on the server, I get > the following lines: > Jun 16 07:45:33 selene perceus-dnsmasq[2854]: DHCPDISCOVER(eth0) > 00:1f:29:ea:e1:ba > Jun 16 07:45:33 selene perceus-dnsmasq[2854]: DHCPOFFER(eth0) > 10.254.0.12 00:1f:29:ea:e1:ba > Jun 16 07:45:37 selene perceus-dnsmasq[2854]: DHCPREQUEST(eth0) > 10.254.0.12 00:1f:29:ea:e1:ba > Jun 16 07:45:37 selene perceus-dnsmasq[2854]: DHCPACK(eth0) 10.254.0.12 > 00:1f:29:ea:e1:ba node02 > Any ideas? > > Thanks, > Olivier > > PS: > the keyboard problems are both in the xfce gui & from the command line, > but all is peachy if I ssh in > from another machine. > > > > Greg Kurtzer wrote: > >> Check to make sure that NFS started up properly >> >> # /etc/init.d/nfs status >> # /etc/init.d/nfs restart >> >> Also, check out the iptables ruleset >> >> # iptables -L >> >> and stop IPTables if necessary for testing: >> >> # /etc/init.d/iptables stop >> # /sbin/chkconfig iptables off >> >> Is the keyboard having issues in the text based virtual terminals, or >> the graphical environment? >> >> Good luck! >> >> >> On Mon, Jun 8, 2009 at 8:05 AM, Olivier >> Marsden wrote: >> >> >>> I'm not sure exactly what caos uses in the firewall domain, but I was >>> supposing that it would at least not interfere with perceus? >>> I'll look into it, thanks for the suggestion. >>> Olivier >>> >>> Scott Kirkby wrote: >>> >>> >>>> Oliver, >>>> >>>> Do you have a firewall on the server interface? That may be block the >>>> nfs mount. >>>> >>>> Scott >>>> >>>> Olivier Marsden wrote: >>>> >>>> >>>> >>>>> Dear Caos users, >>>>> >>>>> I am trying to install a small test cluster using caos+perceus (both >>>>> latest versions), and >>>>> am having problems provisioning the nodes. >>>>> Master node installation went fine, and everything seems to work well >>>>> (except for the keymap for my french keyboard, on which neither shift keys >>>>> nor the alt-gr key work, but that's another question!). >>>>> Sidekick configured perceus for me, if I understood correctly. >>>>> When I boot a node, it correctly gets to the perceus boot menu, but >>>>> then fails to provision from the master (specifically, fails to >>>>> download the vnfs), >>>>> with the following error: >>>>> mount : 10.254.0.1:/var/lib/perceus failed, reason given by server: >>>>> Permission denied >>>>> >>>>> The /etc/exports on the master node seems to be correct, and nfsd is >>>>> running (see end >>>>> of message), and the nic MAC address of my node appears in >>>>> /etc/perceus/ethers . >>>>> Has anyone any idea of what I might be doing incorrectly? >>>>> Thanks in advance! >>>>> >>>>> Olivier Marsden >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> >> cat /etc/exports >>>>> /var/lib/perceus 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # >>>>> perceus export >>>>> /home 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>>>> /usr/local 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus >>>>> export >>>>> /usr/cports 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus >>>>> export >>>>> /opt 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus export >>>>> /srv 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>>>> /private 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>>>> >>>>> >>>>> >> /usr/sbin/rpcinfo -p >>>>> program vers proto port >>>>> 100000 2 tcp 111 portmapper >>>>> 100000 2 udp 111 portmapper >>>>> 100011 1 udp 864 rquotad >>>>> 100011 2 udp 864 rquotad >>>>> 100011 1 tcp 867 rquotad >>>>> 100011 2 tcp 867 rquotad >>>>> 100003 2 udp 2049 nfs >>>>> 100003 3 udp 2049 nfs >>>>> 100003 4 udp 2049 nfs >>>>> 100021 1 udp 45615 nlockmgr >>>>> 100021 3 udp 45615 nlockmgr >>>>> 100021 4 udp 45615 nlockmgr >>>>> 100021 1 tcp 48041 nlockmgr >>>>> 100021 3 tcp 48041 nlockmgr >>>>> 100021 4 tcp 48041 nlockmgr >>>>> 100003 2 tcp 2049 nfs >>>>> 100003 3 tcp 2049 nfs >>>>> 100003 4 tcp 2049 nfs >>>>> 100005 1 udp 59768 mountd >>>>> 100005 1 tcp 55015 mountd >>>>> 100005 2 udp 59768 mountd >>>>> 100005 2 tcp 55015 mountd >>>>> 100005 3 udp 59768 mountd >>>>> 100005 3 tcp 55015 mountd >>>>> 100024 1 udp 57172 status >>>>> 100024 1 tcp 59417 status >>>>> >>>>> _______________________________________________ >>>>> cAos mailing list >>>>> cAos at caoslinux.org >>>>> http://lists.caosity.org/mailman/listinfo/caos >>>>> >>>>> >>>>> >>>>> >>>> _______________________________________________ >>>> cAos mailing list >>>> cAos at caoslinux.org >>>> http://lists.caosity.org/mailman/listinfo/caos >>>> >>>> >>>> >>>> >>> _______________________________________________ >>> cAos mailing list >>> cAos at caoslinux.org >>> http://lists.caosity.org/mailman/listinfo/caos >>> >>> >>> >> >> >> > > _______________________________________________ > cAos mailing list > cAos at caoslinux.org > http://lists.caosity.org/mailman/listinfo/caos > > From gmkurtzer at gmail.com Fri Jun 19 07:33:30 2009 From: gmkurtzer at gmail.com (Greg Kurtzer) Date: Fri, 19 Jun 2009 07:33:30 -0700 Subject: [cAos] problem booting caos/perceus nodes In-Reply-To: <4A3B713D.1090305@ec-lyon.fr> References: <4A2D03EC.6090805@ec-lyon.fr> <4A2D0D96.3000606@etsu.edu> <4A2D28C4.3000403@ec-lyon.fr> <571f1a060906081019x27af0ba6uf6c83e40541fd74c@mail.gmail.com> <4A373571.3070308@ec-lyon.fr> <4A3B713D.1090305@ec-lyon.fr> Message-ID: <571f1a060906190733g51c76b06n9ce7a7003aae2ac7@mail.gmail.com> Hi Oliver, When you checked, was it from a different node or the same node using the same route to the master? (just trying to rule out a bad path) The IP address that the nodes are getting are via your host file. dnsmasq (DHCP/DNS implementation) are grabbing that and statically assigning. Did you set the keyboard type using sidekick? What if you use the "loadkeys" command by hand? Good luck! Greg On Fri, Jun 19, 2009 at 4:06 AM, Olivier Marsden wrote: > Hello again, and sorry to mug people with my problems! > I have tried to test the tftp server by booting a node on > harddisk, configuring it with a dhcp-obtained ip , > and downloading pxelinux.0 by hand with tftp > from the master : it works. > This seems to indicate that the problem comes from the > set-up of the nic card on the node: do people agree with this? > > Thanks, > Olivier > > Olivier Marsden wrote: >> Right, after a few days, I got some time to look into it again. >> And like any self-respecting researcher, I decided that changing >> only one parameter at a time was boring :-) >> So, with an updated system (caos updated via smart), I now get >> different problems, surprise surprise! I managed to boot up the >> nodes once, apparently correctly, and now they get stuck at the >> tftp phase with a tftp open timeout. The nodes receive an ip >> address from the dchp server, in fact the address that I specified >> in the ipaddr module. In my messages log on the server, I get >> the following lines: >> Jun 16 07:45:33 selene perceus-dnsmasq[2854]: DHCPDISCOVER(eth0) >> 00:1f:29:ea:e1:ba >> Jun 16 07:45:33 selene perceus-dnsmasq[2854]: DHCPOFFER(eth0) >> 10.254.0.12 00:1f:29:ea:e1:ba >> Jun 16 07:45:37 selene perceus-dnsmasq[2854]: DHCPREQUEST(eth0) >> 10.254.0.12 00:1f:29:ea:e1:ba >> Jun 16 07:45:37 selene perceus-dnsmasq[2854]: DHCPACK(eth0) 10.254.0.12 >> 00:1f:29:ea:e1:ba node02 >> Any ideas? >> >> Thanks, >> Olivier >> >> PS: >> the keyboard problems are both in the xfce gui & from the command line, >> but all is peachy if I ssh in >> from another machine. >> >> >> >> Greg Kurtzer wrote: >> >>> Check to make sure that NFS started up properly >>> >>> # /etc/init.d/nfs status >>> # /etc/init.d/nfs restart >>> >>> Also, check out the iptables ruleset >>> >>> # iptables -L >>> >>> and stop IPTables if necessary for testing: >>> >>> # /etc/init.d/iptables stop >>> # /sbin/chkconfig iptables off >>> >>> Is the keyboard having issues in the text based virtual terminals, or >>> the graphical environment? >>> >>> Good luck! >>> >>> >>> On Mon, Jun 8, 2009 at 8:05 AM, Olivier >>> Marsden wrote: >>> >>> >>>> I'm not sure exactly what caos uses in the firewall domain, but I was >>>> supposing that it would at least not interfere with perceus? >>>> I'll look into it, thanks for the suggestion. >>>> Olivier >>>> >>>> Scott Kirkby wrote: >>>> >>>> >>>>> Oliver, >>>>> >>>>> Do you have a firewall on the server interface? That may be block the >>>>> nfs mount. >>>>> >>>>> Scott >>>>> >>>>> Olivier Marsden wrote: >>>>> >>>>> >>>>> >>>>>> Dear Caos users, >>>>>> >>>>>> I am trying to install a small test cluster using caos+perceus (both >>>>>> latest versions), and >>>>>> am having problems provisioning the nodes. >>>>>> Master node installation went fine, and everything seems to work well >>>>>> (except for the keymap for my french keyboard, on which neither shift keys >>>>>> ?nor the alt-gr key work, but that's another question!). >>>>>> Sidekick configured perceus for me, if I understood correctly. >>>>>> When I boot a node, it correctly gets to the perceus boot menu, but >>>>>> then ?fails to provision from the master (specifically, fails to >>>>>> download the vnfs), >>>>>> with the following error: >>>>>> mount : 10.254.0.1:/var/lib/perceus failed, reason given by server: >>>>>> Permission denied >>>>>> >>>>>> The /etc/exports on the master node seems to be correct, and nfsd is >>>>>> running (see end >>>>>> of message), and the nic MAC address of my node appears in >>>>>> /etc/perceus/ethers . >>>>>> Has anyone any idea of what I might be doing incorrectly? >>>>>> Thanks in advance! >>>>>> >>>>>> Olivier Marsden >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> ?>> cat /etc/exports >>>>>> /var/lib/perceus 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # >>>>>> perceus export >>>>>> /home 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>>>>> /usr/local 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus >>>>>> export >>>>>> /usr/cports 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus >>>>>> export >>>>>> /opt 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus export >>>>>> /srv 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>>>>> /private 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>>>>> >>>>>> >>>>>> ?>> /usr/sbin/rpcinfo -p >>>>>> ? ?program vers proto ? port >>>>>> ? ? 100000 ? ?2 ? tcp ? ?111 ?portmapper >>>>>> ? ? 100000 ? ?2 ? udp ? ?111 ?portmapper >>>>>> ? ? 100011 ? ?1 ? udp ? ?864 ?rquotad >>>>>> ? ? 100011 ? ?2 ? udp ? ?864 ?rquotad >>>>>> ? ? 100011 ? ?1 ? tcp ? ?867 ?rquotad >>>>>> ? ? 100011 ? ?2 ? tcp ? ?867 ?rquotad >>>>>> ? ? 100003 ? ?2 ? udp ? 2049 ?nfs >>>>>> ? ? 100003 ? ?3 ? udp ? 2049 ?nfs >>>>>> ? ? 100003 ? ?4 ? udp ? 2049 ?nfs >>>>>> ? ? 100021 ? ?1 ? udp ?45615 ?nlockmgr >>>>>> ? ? 100021 ? ?3 ? udp ?45615 ?nlockmgr >>>>>> ? ? 100021 ? ?4 ? udp ?45615 ?nlockmgr >>>>>> ? ? 100021 ? ?1 ? tcp ?48041 ?nlockmgr >>>>>> ? ? 100021 ? ?3 ? tcp ?48041 ?nlockmgr >>>>>> ? ? 100021 ? ?4 ? tcp ?48041 ?nlockmgr >>>>>> ? ? 100003 ? ?2 ? tcp ? 2049 ?nfs >>>>>> ? ? 100003 ? ?3 ? tcp ? 2049 ?nfs >>>>>> ? ? 100003 ? ?4 ? tcp ? 2049 ?nfs >>>>>> ? ? 100005 ? ?1 ? udp ?59768 ?mountd >>>>>> ? ? 100005 ? ?1 ? tcp ?55015 ?mountd >>>>>> ? ? 100005 ? ?2 ? udp ?59768 ?mountd >>>>>> ? ? 100005 ? ?2 ? tcp ?55015 ?mountd >>>>>> ? ? 100005 ? ?3 ? udp ?59768 ?mountd >>>>>> ? ? 100005 ? ?3 ? tcp ?55015 ?mountd >>>>>> ? ? 100024 ? ?1 ? udp ?57172 ?status >>>>>> ? ? 100024 ? ?1 ? tcp ?59417 ?status >>>>>> >>>>>> _______________________________________________ >>>>>> cAos mailing list >>>>>> cAos at caoslinux.org >>>>>> http://lists.caosity.org/mailman/listinfo/caos >>>>>> >>>>>> >>>>>> >>>>>> >>>>> _______________________________________________ >>>>> cAos mailing list >>>>> cAos at caoslinux.org >>>>> http://lists.caosity.org/mailman/listinfo/caos >>>>> >>>>> >>>>> >>>>> >>>> _______________________________________________ >>>> cAos mailing list >>>> cAos at caoslinux.org >>>> http://lists.caosity.org/mailman/listinfo/caos >>>> >>>> >>>> >>> >>> >>> >> >> _______________________________________________ >> cAos mailing list >> cAos at caoslinux.org >> http://lists.caosity.org/mailman/listinfo/caos >> >> > > _______________________________________________ > cAos mailing list > cAos at caoslinux.org > http://lists.caosity.org/mailman/listinfo/caos > -- Greg Kurtzer http://www.infiscale.com/ http://www.perceus.org/ http://www.caoslinux.org/ From Olivier.Marsden at ec-lyon.fr Fri Jun 19 09:18:13 2009 From: Olivier.Marsden at ec-lyon.fr (Olivier Marsden) Date: Fri, 19 Jun 2009 18:18:13 +0200 Subject: [cAos] problem booting caos/perceus nodes In-Reply-To: <571f1a060906190733g51c76b06n9ce7a7003aae2ac7@mail.gmail.com> References: <4A2D03EC.6090805@ec-lyon.fr> <4A2D0D96.3000606@etsu.edu> <4A2D28C4.3000403@ec-lyon.fr> <571f1a060906081019x27af0ba6uf6c83e40541fd74c@mail.gmail.com> <4A373571.3070308@ec-lyon.fr> <4A3B713D.1090305@ec-lyon.fr> <571f1a060906190733g51c76b06n9ce7a7003aae2ac7@mail.gmail.com> Message-ID: <4A3BBA45.3040504@ec-lyon.fr> Hi Greg, I checked from the same node with the same route, ie I manually assigned it the ip address it was receiving by dhcp upon auto boot. The keyboard on the master was indeed set using sidekick. I don't know the "loadkeys" command, so will look into that one at some point. Thanks for your help! Olivier Greg Kurtzer wrote: > Hi Oliver, > > When you checked, was it from a different node or the same node using > the same route to the master? (just trying to rule out a bad path) > > The IP address that the nodes are getting are via your host file. > dnsmasq (DHCP/DNS implementation) are grabbing that and statically > assigning. > > Did you set the keyboard type using sidekick? What if you use the > "loadkeys" command by hand? > > Good luck! > Greg > > > > On Fri, Jun 19, 2009 at 4:06 AM, Olivier > Marsden wrote: > >> Hello again, and sorry to mug people with my problems! >> I have tried to test the tftp server by booting a node on >> harddisk, configuring it with a dhcp-obtained ip , >> and downloading pxelinux.0 by hand with tftp >> from the master : it works. >> This seems to indicate that the problem comes from the >> set-up of the nic card on the node: do people agree with this? >> >> Thanks, >> Olivier >> >> Olivier Marsden wrote: >> >>> Right, after a few days, I got some time to look into it again. >>> And like any self-respecting researcher, I decided that changing >>> only one parameter at a time was boring :-) >>> So, with an updated system (caos updated via smart), I now get >>> different problems, surprise surprise! I managed to boot up the >>> nodes once, apparently correctly, and now they get stuck at the >>> tftp phase with a tftp open timeout. The nodes receive an ip >>> address from the dchp server, in fact the address that I specified >>> in the ipaddr module. In my messages log on the server, I get >>> the following lines: >>> Jun 16 07:45:33 selene perceus-dnsmasq[2854]: DHCPDISCOVER(eth0) >>> 00:1f:29:ea:e1:ba >>> Jun 16 07:45:33 selene perceus-dnsmasq[2854]: DHCPOFFER(eth0) >>> 10.254.0.12 00:1f:29:ea:e1:ba >>> Jun 16 07:45:37 selene perceus-dnsmasq[2854]: DHCPREQUEST(eth0) >>> 10.254.0.12 00:1f:29:ea:e1:ba >>> Jun 16 07:45:37 selene perceus-dnsmasq[2854]: DHCPACK(eth0) 10.254.0.12 >>> 00:1f:29:ea:e1:ba node02 >>> Any ideas? >>> >>> Thanks, >>> Olivier >>> >>> PS: >>> the keyboard problems are both in the xfce gui & from the command line, >>> but all is peachy if I ssh in >>> from another machine. >>> >>> >>> >>> Greg Kurtzer wrote: >>> >>> >>>> Check to make sure that NFS started up properly >>>> >>>> # /etc/init.d/nfs status >>>> # /etc/init.d/nfs restart >>>> >>>> Also, check out the iptables ruleset >>>> >>>> # iptables -L >>>> >>>> and stop IPTables if necessary for testing: >>>> >>>> # /etc/init.d/iptables stop >>>> # /sbin/chkconfig iptables off >>>> >>>> Is the keyboard having issues in the text based virtual terminals, or >>>> the graphical environment? >>>> >>>> Good luck! >>>> >>>> >>>> On Mon, Jun 8, 2009 at 8:05 AM, Olivier >>>> Marsden wrote: >>>> >>>> >>>> >>>>> I'm not sure exactly what caos uses in the firewall domain, but I was >>>>> supposing that it would at least not interfere with perceus? >>>>> I'll look into it, thanks for the suggestion. >>>>> Olivier >>>>> >>>>> Scott Kirkby wrote: >>>>> >>>>> >>>>> >>>>>> Oliver, >>>>>> >>>>>> Do you have a firewall on the server interface? That may be block the >>>>>> nfs mount. >>>>>> >>>>>> Scott >>>>>> >>>>>> Olivier Marsden wrote: >>>>>> >>>>>> >>>>>> >>>>>> >>>>>>> Dear Caos users, >>>>>>> >>>>>>> I am trying to install a small test cluster using caos+perceus (both >>>>>>> latest versions), and >>>>>>> am having problems provisioning the nodes. >>>>>>> Master node installation went fine, and everything seems to work well >>>>>>> (except for the keymap for my french keyboard, on which neither shift keys >>>>>>> nor the alt-gr key work, but that's another question!). >>>>>>> Sidekick configured perceus for me, if I understood correctly. >>>>>>> When I boot a node, it correctly gets to the perceus boot menu, but >>>>>>> then fails to provision from the master (specifically, fails to >>>>>>> download the vnfs), >>>>>>> with the following error: >>>>>>> mount : 10.254.0.1:/var/lib/perceus failed, reason given by server: >>>>>>> Permission denied >>>>>>> >>>>>>> The /etc/exports on the master node seems to be correct, and nfsd is >>>>>>> running (see end >>>>>>> of message), and the nic MAC address of my node appears in >>>>>>> /etc/perceus/ethers . >>>>>>> Has anyone any idea of what I might be doing incorrectly? >>>>>>> Thanks in advance! >>>>>>> >>>>>>> Olivier Marsden >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >> cat /etc/exports >>>>>>> /var/lib/perceus 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # >>>>>>> perceus export >>>>>>> /home 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>>>>>> /usr/local 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus >>>>>>> export >>>>>>> /usr/cports 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus >>>>>>> export >>>>>>> /opt 10.254.0.0/255.255.252.0(ro,no_root_squash,async) # perceus export >>>>>>> /srv 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>>>>>> /private 10.254.0.0/255.255.252.0(rw,no_root_squash,async) # perceus export >>>>>>> >>>>>>> >>>>>>> >> /usr/sbin/rpcinfo -p >>>>>>> program vers proto port >>>>>>> 100000 2 tcp 111 portmapper >>>>>>> 100000 2 udp 111 portmapper >>>>>>> 100011 1 udp 864 rquotad >>>>>>> 100011 2 udp 864 rquotad >>>>>>> 100011 1 tcp 867 rquotad >>>>>>> 100011 2 tcp 867 rquotad >>>>>>> 100003 2 udp 2049 nfs >>>>>>> 100003 3 udp 2049 nfs >>>>>>> 100003 4 udp 2049 nfs >>>>>>> 100021 1 udp 45615 nlockmgr >>>>>>> 100021 3 udp 45615 nlockmgr >>>>>>> 100021 4 udp 45615 nlockmgr >>>>>>> 100021 1 tcp 48041 nlockmgr >>>>>>> 100021 3 tcp 48041 nlockmgr >>>>>>> 100021 4 tcp 48041 nlockmgr >>>>>>> 100003 2 tcp 2049 nfs >>>>>>> 100003 3 tcp 2049 nfs >>>>>>> 100003 4 tcp 2049 nfs >>>>>>> 100005 1 udp 59768 mountd >>>>>>> 100005 1 tcp 55015 mountd >>>>>>> 100005 2 udp 59768 mountd >>>>>>> 100005 2 tcp 55015 mountd >>>>>>> 100005 3 udp 59768 mountd >>>>>>> 100005 3 tcp 55015 mountd >>>>>>> 100024 1 udp 57172 status >>>>>>> 100024 1 tcp 59417 status >>>>>>> >>>>>>> _______________________________________________ >>>>>>> cAos mailing list >>>>>>> cAos at caoslinux.org >>>>>>> http://lists.caosity.org/mailman/listinfo/caos >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>> _______________________________________________ >>>>>> cAos mailing list >>>>>> cAos at caoslinux.org >>>>>> http://lists.caosity.org/mailman/listinfo/caos >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>> _______________________________________________ >>>>> cAos mailing list >>>>> cAos at caoslinux.org >>>>> http://lists.caosity.org/mailman/listinfo/caos >>>>> >>>>> >>>>> >>>>> >>>> >>>> >>> _______________________________________________ >>> cAos mailing list >>> cAos at caoslinux.org >>> http://lists.caosity.org/mailman/listinfo/caos >>> >>> >>> >> _______________________________________________ >> cAos mailing list >> cAos at caoslinux.org >> http://lists.caosity.org/mailman/listinfo/caos >> >> > > > > From Olivier.Marsden at ec-lyon.fr Tue Jun 30 08:10:50 2009 From: Olivier.Marsden at ec-lyon.fr (Olivier Marsden) Date: Tue, 30 Jun 2009 17:10:50 +0200 Subject: [cAos] problem booting caos/perceus nodes Message-ID: <4A4A2AFA.9090404@ec-lyon.fr> Hello again, I am slowing getting to the root of my problems, and it appears to be a hardware/firmware/bios problem. Indeed, consecutive reboots of the same node yield flip-of-the-coin type results for successful pxe-booting or not, apparently totally at random. This is with the latest firmware for Broadcom netextreme cards on HP DL165 G5 hardware. So, with a little bit of persistence, I am now able to boot all the nodes correctly! I now have two follow-up questions: - I would like the nic cards (two dual-port cards per node) to be named consistently, which doesn't happen for the moment; I know how to do this with udev and 70-persistent-net.rules on a harddisk install, but have no clue for a diskless install? - I am using a scratch partition on local hard disks. I have put a line in the fstab in the vnfs, with the correct device name etc, and made the mountpoint in the vnfs file system, but the mount is not performed during the diskless install. What am I missing here? (sorry if this is more of a perceus question; should I post it there?) The overall architecture of caos+perceus, now that I'm getting more familiar with it, looks really elegant. Many thanks on making software like this available to the unwashed masses! Olivier Marsden From gmkurtzer at gmail.com Tue Jun 30 12:41:25 2009 From: gmkurtzer at gmail.com (Greg Kurtzer) Date: Tue, 30 Jun 2009 12:41:25 -0700 Subject: [cAos] problem booting caos/perceus nodes In-Reply-To: <4A4A2AFA.9090404@ec-lyon.fr> References: <4A4A2AFA.9090404@ec-lyon.fr> Message-ID: <571f1a060906301241wcb82249w22f6bd9b12c31a1f@mail.gmail.com> On Tue, Jun 30, 2009 at 8:10 AM, Olivier Marsden wrote: > Hello again, > > I am slowing getting to the root of my problems, and it appears to be a > hardware/firmware/bios problem. > Indeed, consecutive reboots of the same node yield flip-of-the-coin type > results for successful pxe-booting > or not, apparently totally at random. This is with the latest firmware > for Broadcom netextreme cards on HP > DL165 G5 hardware. > So, with a little bit of persistence, I am now able to boot all the > nodes correctly! > > I now have two follow-up questions: > - I would like the nic cards (two dual-port cards per node) to be named > consistently, which doesn't happen for the > ?moment; I know how to do this with udev and 70-persistent-net.rules on > a harddisk install, but have no clue for > ?a diskless install? What do you mean they are not consistent? Like on one boot, nic1 = eth0 and on the second nic2 = eth0? Do they use the same kernel module/driver? You can always create a nodescript to add the correct entries for udev, but it sounds like something else weirder is going on. > - I am using a scratch partition on local hard disks. I have put a line > in the fstab in the vnfs, with the correct device name > etc, and made the mountpoint in the vnfs file system, but the mount is > not performed during the diskless install. What am > ?I missing here? (sorry if this is more of a perceus question; should I > post it there?) Try to add "mount -a" to the /etc/rc.local in the VNFS. It has to do with the fact that /fastboot exists which bypasses local mounts. We already have this listed as a bug and are undergoing testing of solutions now. > > The overall architecture of caos+perceus, now that I'm getting more > familiar with it, looks really elegant. > Many thanks on making software like this available to the unwashed masses! > Olivier Marsden Thanks for the thanks! It is always great to get feedback from our user community, especially when it is good. :) Greg -- Greg Kurtzer http://www.infiscale.com/ http://www.perceus.org/ http://www.caoslinux.org/