[ipxe-devel] [Issue]Latest iPXE leads to elilo hang

Zejun ZJ2 Chen chenzj2 at lenovo.com
Wed Nov 7 08:52:23 UTC 2018


Hello iPXE team,

Lenovo DCG team has used iPXE for years,  recently we encountered one critical issue when we were using the latest iPXE code.  As Li Na mentioned below,  the issue doesn't exist in the specified older versions. Let me try to summarize our code working flow and the symptom as below.

1: x86 Server boot  ->  2: UEFI  ->  3: iPXE -> 4: Linux elilo bootloader -> 5: Linux Ramdisk OS as OS  installer.

The steps #1 to #3 work properly,  installation stopped between step #4 and #5.  We enabled elilo debug mode, and caught one error message  'util.c (line 134): waitkey  WaitForEvent failed unsupported'.  Please refer to the attached screen shot.


We use below iPXE command to download elilo via network and run it. The attached xxx.elilo file figures out the Linux kernel and ramfs  files' location in network.

:retryelilo
chain --timeout 60000 https://10.240.210.114:8443/tftpboot/xcat/elilo-x64.efi -C /tftpboot/xcat/xnba/nodes/node9AFA2DDBDE6.elilo || goto retryelilo

Refer to the messages, elilo successfully downloaded kernel and ramfs files via network, but threw a critical error when elilo were installing Linux kernel .

Could your team help to investigate and solve the issue?

Thanks,

Abraham Chen


From: Na NL Li
Sent: Friday, November 2, 2018 3:32 PM
To: mbrown at fensystems.co.uk
Cc: Zejun ZJ2 Chen; Joseph Sankar
Subject: [Issue]Latest iPXE leads to elilo hang

Hello Michael,

I'm Li Na from Lenovo xClarity team, working on OS deployment component. Our team had contacted with you before to enable IPv6, and now we're planning to upgrade ipxe to the latest. But during UT, we've found some issue:
With latest ipxe source code, we built new image to boot elilo and then load initrd for linux, but it always hangs after loading is done. With the error code 2a0860 <http://ipxe.org/err/2a0860> on the screen, it points to "Error: No buffer space available". And this out-of-space error seems also explaining why ramdisk fails to run. So is there any recent ipxe code change may lead to this issue? Or what could we do to make this latest ipxe work for us?

Thanks for your help in advance!

BTW, we're currently using commit 8dbb73a779e5b11ee2b65f9d2af6dd9bd8998608<https://git.ipxe.org/ipxe.git/commit/8dbb73a779e5b11ee2b65f9d2af6dd9bd8998608> on Feb 20th, 2018. This one works well.

[cid:image001.png at 01D476B1.AB890680]
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ipxe.org/pipermail/ipxe-devel/attachments/20181107/f9ff93f4/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 57217 bytes
Desc: image001.png
URL: <http://lists.ipxe.org/pipermail/ipxe-devel/attachments/20181107/f9ff93f4/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: elilo-hang-debug.png
Type: image/png
Size: 46773 bytes
Desc: elilo-hang-debug.png
URL: <http://lists.ipxe.org/pipermail/ipxe-devel/attachments/20181107/f9ff93f4/attachment-0001.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: node9AFA2DDBDE6.elilo
Type: application/octet-stream
Size: 488 bytes
Desc: node9AFA2DDBDE6.elilo
URL: <http://lists.ipxe.org/pipermail/ipxe-devel/attachments/20181107/f9ff93f4/attachment.obj>


More information about the ipxe-devel mailing list