https://bugzilla.novell.com/show_bug.cgi?id=459118 Summary: kdump failed to to trigger dump on network through NFS in openSuSE 11.1 Beta5-ppc Product: openSUSE 11.1 Version: RC 1 Platform: PowerPC-64 OS/Version: All Status: NEW Severity: Major Priority: P5 - None Component: Kernel AssignedTo: bnc-team-screening@forge.provo.novell.com ReportedBy: bugproxy@us.ibm.com QAContact: qa@suse.de Found By: Third Party Developer/Partner Partner ID: LTC 50212 =Comment: #0================================================= Pavan Naregundi <pavan.naregundi@in.ibm.com> - In openSuSE 11.1 Beta5 while triggering kdump on network through NFS following error message was produced =============================== # echo c > /proc/sysrq-trigger SysRq : Trigger a crashdump Sending IPI to other cpus... doing fast boot SysRq : Changing Loglevel Loglevel set to 1 Creating device nodes with udev blogd: can not open pty/tty pair: No such file or directory [NETWORK] using static config based on ip=9.124.111.249::9.124.111.1:255.255.255.0:llm62:eth0:none Ignoring mode none, using static configuration Cannot find device "eth0" Cannot find device "eth0" RTNETLINK answers: No such process Waiting for device /dev/sda6 to appear: ok fsck 1.41.1 (01-Sep-2008) [/sbin/fsck.ext3 (1) -- /] fsck.ext3 -a /dev/sda6 /dev/sda6: recovering journal /dev/sda6: Clearing orphaned inode 378564 (uid=0, gid=0, mode=0100600, size=217016) /dev/sda6: clean, 177255/1313280 files, 2000377/5242880 blocks fsck succeeded. Mounting root device read-write. Mounting root /dev/sda6 Deletion of old dump only on local disk. Cannot execute showmount --directories 9.124.111.97:showmount: RPC: Unable to send Last command failed (255). bash-3.2# ============================== KDUMP_SAVEDIR="nfs://9.124.111.97/var/crash" Remote directory on nfs server was properly exported and could be mounted manually # cat /etc/exports /var/crash *(rw,no_root_squash,sync) # Machine : llm62 CPU Type : Power 6 =Comment: #1================================================= Nagesh Sharyathi <sharyath@in.ibm.com> - Pavan Can you provide the machine access Thanks Sharyathi =Comment: #2================================================= Pavan Naregundi <pavan.naregundi@in.ibm.com> - (In reply to comment #1) Provided machine access. =Comment: #3================================================= Nagesh Sharyathi <sharyath@in.ibm.com> - The problem is because the kdump kernel is not built with CONFIG_EHEA=m, I will rebuild the kernel and test, than we can mirror it to Novel Thanks Sharyathi =Comment: #4================================================= Nagesh Sharyathi <sharyath@in.ibm.com> - Pavan Can you let me know where I can access kernel source code Thanks Sharyathi =Comment: #6================================================= Nagesh Sharyathi <sharyath@in.ibm.com> - I have built the kernel let me know when I can use it Thanks Sharyathi =Comment: #8================================================= Nagesh Sharyathi <sharyath@in.ibm.com> - I tried rebuilding the kernel with CONFIG_EHEA. It looks like the mkinitrd scripts needs to be updated to included the ehea driver. I am looking into updating the mkinitrd scripts Thanks Sharyathi =Comment: #10================================================= Nagesh Sharyathi <sharyath@in.ibm.com> - /* Excuse me for the partial Update*/ Novel can you please let us know your thoughts on this There are multiple issues with the problem Issue1: kdump kernel not built with ehea driver. I observed that kdump kernel and kdump initrd file haven't enabled CONFIG_EHEA and subsequent boot into kdump kernel fails to configure network driver. Since CONFIG_EHEA has dependency with CONFIG_IBMBUS we need to enable it before enabling CONFIG_EHEA. Issue2: When you select network dump over NFS yast fails to build initrd file with ehea module I tried manually building initrd file: Using mkinitrd -k <vmlinux image> -i <inird image> -f network, still fails to include ehea driver in initrd file. work around is to include ehea driver specifically ie: mkinitrd -k <vmlinux image> -i <inird image> -m ehea Please let me know will this be corrected in yast or should we have to change /lib/mkinitrd/scripts/scripts/setup-network.sh to include ehea driver even when -f network option is provided to the mkinitrd. After building ehea driver I am facing problem with mounting root filesystem. Not sure why the machine is trying to mount root file system. When the option chosen is network dump. Please let us know your thoughts Thanks Sharyathi -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.