I have an intermittent problem with my Dell blades, out of 80 blades 69 of them kickstarted Centos 6 fine using PXE The other 11, I get a c0t0 not found error (indicating it is not finding the local disk on the blade).
I can remote mount the iso image and do a basic install of centos on these blades, but when I go to do a pxe boot it gives me that error again.
I have googled the issue and come up empty handed. Dell told us to install centos 6 minimal ISO then pxe install the server. This has been unsuccessful. I get the local ISO to install but the pxe fails again.
Since the kickstart file defines the disk as sda1 I am supposing that it is really a hardware problem. But I am told by coworkers who handed it off to me that it is a known issue with centos and Dell blades.
These are two internal disks on the blade.
Any suggestions?
Dan Hyatt wrote:
I have an intermittent problem with my Dell blades, out of 80 blades 69 of them kickstarted Centos 6 fine using PXE The other 11, I get a c0t0 not found error (indicating it is not finding the local disk on the blade).
<snip>
Since the kickstart file defines the disk as sda1 I am supposing that it is really a hardware problem.
<snip> Nahhh... First, I suspect that those 11 blades are a slightly different model - newer, or whatever. When we first got some Dell R720s was when we ran into CentOS calling the NIC em1 instead of eth0, while *other* R-series servers we got the same time were still eth0.
I'm afraid you're going to have to put conditionals into the ks, or maybe, the way we do, write a CGI that creates one on the fly, to tell it to look for those drives under those names. *shrug* We did just that in our CGI, and chose CentOS 6, 64 bit, R720 for them.
mark
On Thu, Apr 17, 2014 at 4:04 PM, Dan Hyatt dhyatt@dsgmail.wustl.edu wrote:
I have an intermittent problem with my Dell blades, out of 80 blades 69 of them kickstarted Centos 6 fine using PXE The other 11, I get a c0t0 not found error (indicating it is not finding the local disk on the blade).
I can remote mount the iso image and do a basic install of centos on these blades, but when I go to do a pxe boot it gives me that error again.
I have googled the issue and come up empty handed. Dell told us to install centos 6 minimal ISO then pxe install the server. This has been unsuccessful. I get the local ISO to install but the pxe fails again.
Since the kickstart file defines the disk as sda1 I am supposing that it is really a hardware problem. But I am told by coworkers who handed it off to me that it is a known issue with centos and Dell blades.
These are two internal disks on the blade.
Any suggestions?
install "working" blade from .iso install "non-working" blade from .iso diff anaconda-ks.cfg.working ananconda-ks.cfg.non-working
-
Dan Hyatt -
m.roth＠5-cent.us -
Steven Tardy