Hi, Note, that this post follows the thread "OSCAR : CentOS 6.4/6.5 : error with OSCAR dependencies, image stage". Sorry for the interruption, I was out of my office.
First, I just want give you some feedback with my attempt to install oscar 6 using centOS 6.5 but with local repo (because the internet connection is too slow to make tests). After some minor fix using local repos, I can pull with PXE the image to my client server (Dell poweredge R720xd, same as the golden node) using UYOK. From my notes, my minor problem at this stage (with "tftp timeout" symptoms like from the client side) were due to broken rsync and atftpd (service not alive or with wrong arguments, see below :) ATFTPD : -------- the wizard-oscar.log says : \\ 2014-2-21 16:55:32 [main :: Line 137] Enabling atftpd [INFO] Setting service xinetd to on... [INFO] Called getitem with xinetd_service and returning xinetd [INFO] xinetd is already on [INFO] Setting xinetd service tftp to on... [INFO] Called getitem with tftp_service and returning atftp [ACTION] About to run: /sbin/chkconfig atftp on erreur lors de la lecture d'informations sur le service atftpÂ| : Aucun fichier ou dossier de ce type [ERROR] Failed to run: /sbin/chkconfig atftp on (rc=1) at /usr/share/perl5/vendor_perl/OSCAR/Utils.pm line 101 [INFO] Performing restart on xinetd service. // Using "netstat -an | fgrep -w 69", the port 69 stays closed. Then, after googling, in order to start the service atftpd, I made some change in /etc/sysconfig/atftpd by setting USE_INET from false to true. But the reason remains unclear ... RSYNCD : ------ The second problem was related to rsyncd, with error like : \\ get_scripts_directory rsync -a 10.1.1.1::scripts/ /scripts/ @ERROR: Unknown module 'scripts' rsync error: error starting client-server protocol (code 5) at main.c(1536) [receiver=3.0.0] // The error is related to the service rsyncd. It runs, but by checking with "ps -ef | grep -i rsync", the config argument for rsyncd was wrong : config=/tmp/fileNWWV3 instead of config=/etc/systemimager/rsyncd.conf Then, I stop by hand the rsyncd service and do : ~]# /etc/init.d/systemimager-server-rsyncd start Starting rsync daemon for systemimager: ok. ~]# ps -ef | grep rsync root 6503 1 0 11:29 ? 00:00:00 /usr/bin/rsync --daemon --config=/etc/systemimager/rsyncd.conf The reason remains unclear, but now I can pull the image. MY MAIN QUESTION : ------------------ Now, I m fighting with the grub loader and network setup just after the image installation. What is unclear for me is why some large sections calling "system configurator" are now commented out in Server.pm ? How do you proceed to launch the boot loader using grub-install on the client node ? Always with SystemConfig/Boot/Grub.pm ? But how/where you call it ? Same remark apply with the setting of network NICs on the client node : if I'm not wrong, in previous Oscar version, this stage was performed by calling the system configurator, But, now with systemconfigurator off, how do you proceed to write/change the file /etc/sysconfig/network and icfg-eth0 inside the client server ? These files look unchanged (more precisely, they are missing) and the client node stays unreachable from the cluster network after the image installation ? Many thanks for your comments. Jerome ------------------------------------------------------------------------------ Flow-based real-time traffic analytics software. Cisco certified tool. Monitor traffic, SLAs, QoS, Medianet, WAAS etc. with NetFlow Analyzer Customize your own dashboards, set traffic alerts and generate reports. Network behavioral analysis & security monitoring. All-in-one tool. http://pubads.g.doubleclick.net/gampad/clk?id=126839071&iu=/4140/ostg.clktrk _______________________________________________ Oscar-users mailing list Oscar-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/oscar-users
