Hi,
First of all, I have searched a bit in the forums but for some reason it seems I am not able to find a matching issue, except one that didnt resolve my problem.
I have installed Proxmox according to the instructions here: https://pve.proxmox.com/wiki/Proxmox_ISCSI_installation as well as adding info from a few other places since it assumed dhcp etc.
Right now I have a working Debian system that boots from iSCSI, using the 2.6.32-5-adm64 kernel. The idea was then to use the Proxmox kernel and, going forward, boot into it. So I looked up the guide here: https://pve.proxmox.com/wiki/Install_Proxmox_VE_on_Debian_Wheezy
Adding sources, installing images, updating grub etc works great. No issues at all. However, when I boot the PVE kernel the machine fails to boot due to - what it seems - a disk error. Starting the iSCSI session etc seems to be working based on the messages printed on the screen.
It is quite hard to see exactly what is the first error, since the screen is flooded but at the end I get the error message: ext4-fs (sda2): ext4_da_writepages: jbd2_start: 1024 pages, ino [then a changing number]. this message continues to appear and are pushing out the buffer on the top as well so its quite hard to see what the actual issue is.
An other error messages are "Rejecting IO to offline device". At one point I think I saw (before it was pushed out) that the drives where considered having hardware issues and that I should run fsck.
I've also enabled bootlogd to see all the log messages but that log seems to be very empty unless I boot into the Debian kernel, same goes for dmesg.
I've tried different pve kernel versions, including 2.6.32-25, 2.6.32-26, 2.6.32-27, 2.6.32-34 but all with the same issue. I've ran fsck froma live dist and there was no error detected (it finnished instant tho). However, whenever I try to boot into the Debian kernel, there seems to be no issue with the filesystem.
Any idea of how I can proceed with my troubleshooting and resolve the issue? I understand remote boot is not officially supported, but in this case it might not even be related since the iSCSI session is already up and are working under another kernel. Likely there is some issue with modules that needs to be enabled in the kernel, or at least that is my guess.
Thanks in advance,
trafficlight
First of all, I have searched a bit in the forums but for some reason it seems I am not able to find a matching issue, except one that didnt resolve my problem.
I have installed Proxmox according to the instructions here: https://pve.proxmox.com/wiki/Proxmox_ISCSI_installation as well as adding info from a few other places since it assumed dhcp etc.
Right now I have a working Debian system that boots from iSCSI, using the 2.6.32-5-adm64 kernel. The idea was then to use the Proxmox kernel and, going forward, boot into it. So I looked up the guide here: https://pve.proxmox.com/wiki/Install_Proxmox_VE_on_Debian_Wheezy
Adding sources, installing images, updating grub etc works great. No issues at all. However, when I boot the PVE kernel the machine fails to boot due to - what it seems - a disk error. Starting the iSCSI session etc seems to be working based on the messages printed on the screen.
It is quite hard to see exactly what is the first error, since the screen is flooded but at the end I get the error message: ext4-fs (sda2): ext4_da_writepages: jbd2_start: 1024 pages, ino [then a changing number]. this message continues to appear and are pushing out the buffer on the top as well so its quite hard to see what the actual issue is.
An other error messages are "Rejecting IO to offline device". At one point I think I saw (before it was pushed out) that the drives where considered having hardware issues and that I should run fsck.
I've also enabled bootlogd to see all the log messages but that log seems to be very empty unless I boot into the Debian kernel, same goes for dmesg.
I've tried different pve kernel versions, including 2.6.32-25, 2.6.32-26, 2.6.32-27, 2.6.32-34 but all with the same issue. I've ran fsck froma live dist and there was no error detected (it finnished instant tho). However, whenever I try to boot into the Debian kernel, there seems to be no issue with the filesystem.
Any idea of how I can proceed with my troubleshooting and resolve the issue? I understand remote boot is not officially supported, but in this case it might not even be related since the iSCSI session is already up and are working under another kernel. Likely there is some issue with modules that needs to be enabled in the kernel, or at least that is my guess.
Thanks in advance,
trafficlight