Red Hat Bugzilla – Bug 57379
installer errors out during net install
Last modified: 2007-04-18 12:38:39 EDT
From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; U; WinNT4.0; en-US; rv:0.9.6+) Gecko/20011206 Description of problem: I was doing an net install to my laptop (a Compaq Armarda m300 which lacks a cdrom drive) from another 7.2 box set up as an NFS server. I created the boot disk as required, and start the installer okay. When it gets up to the installing packages stage, it stops with an error after a few packages, dropping back to the console with an error message (install exited abnormally -- received signal 11) Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1.start a network install using the bootnet.img floppy image 2.answer all questions in installler to install from NFS server 3.wait for installer to fail during package installation Actual Results: Installer stops. Sometimes it stays in graphical mode (in which case, ctrl+alt+F1 gets back to the console), and sometimes it drops back to text mode itself. Expected Results: Installation of selected packages to complete successfully. Additional info: on VT1, I get the following message: install exited abnormally -- received signal 11 it then unmounts the volumes and says it is safe to reboot. once I also got the following message before the first: libpng error: invalid stored block lengths I checked on VT4, and there are lots of messages like: <4>Error -3 while decompressing! <4>c0266c1b(1)->c732b000(4096) (the numbers are different each time). The system was running RH6.2 without problem before the install (I chose to reformat the linux partitions as part of the install, so it is a clean install).
I borrowed a base for the laptop with a cdrom drive, so it looks like I can finish the install. My laptop was using the builtin EtherExpress Pro 100 for the network install, and the desktop system acting as the NFS server was running without problem. Both machines were configured via DHCP, so had correct network settings. I was not able to see what message occured on VT4 before the decompression error messages.
This looks very similar to bug 57425. I can only speculate that the install files copied onto the NFS server were corrupted by the copy. Did you check the md5sum of the files against the original source?
I had the first cd mounted as /mnt/cdrom on the nfs server, and added it directly to the exports list. I had already installed a few systems off that cd, so I don't think the files were corrupt. I don't think the boot floppy was a problem either -- if it was, the problem would have occured earlier.
There is now a resolution to bug #57425. Can you look at that report and see if that fix could help you? Try booting with 'linux ramdisk_size=7198'. Does that help?
Closing due to inactivity. Please reopen if you have more information.