Created attachment 478354 [details] system halt screenshot Description of problem: After kickstart installation, system halts after first reboot Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. kickstart install F-15 2. reboot 3. Actual results: Expected results: Additional info:
Created attachment 478356 [details] anaconda log
Created attachment 478357 [details] install log
Created attachment 478358 [details] anaconda program log
Created attachment 478359 [details] anaconda syslog
Created attachment 478360 [details] install syslog
Yeah, I can reproduce it as well using kickstart to install F15-alpha-tc1. It happens when clicking 'Reboot button' after package installation finished.
What's your kickstart file look like?
Created attachment 478537 [details] kickstart configuration
Halting is the default action for kickstart when nothing else is specified, though: https://fedoraproject.org/wiki/Anaconda/Kickstart#halt
I'm experiencing the same issue, but I have 'reboot' in my kickstart file. I was doing a cmdline install test, so I'm not able to grab any logs off the system. I'll see what I can uncover ... # Firewall configuration firewall --disabled # Use network installation url --url="http://dell-t5400.test.redhat.com/cobbler/ks_mirror/F-15-Alpha-x86_64" # Root password rootpw --iscrypted $1$xafj7qlW$6swjxMwu0po47drJVRcIZ/ # System keyboard keyboard us # System language lang en_US # Installation logging level logging --level=info # Reboot after installation reboot # System timezone timezone --isUtc America/New_York # System bootloader configuration bootloader --location=mbr autopart # Partition clearing information clearpart --all --initlabel %packages --default --ignoremissing %end
reproduced with kickstart file #install a fresh system rather than upgrade an existing system install # Perform the kickstart installation in text mode #text graphical # Install from an installation tree on a remote server via FTP or HTTP cdrom #fire wall configuration firewall --disabled # Root password rootpw `````` # System keyboard keyboard us # System language lang en_US # Installation logging level logging --level=info # Reboot after installation reboot # System timezone timezone --isUtc America/New_York # System bootloader configuration bootloader --location=mbr autopart # Partition clearing information clearpart --all --initlabel %packages --default --ignoremissing %end
I think this should block the beta (or at least the final release) because it breaks the full interactivity of a kickstart install.
At the least, this should be fixed in anaconda-16.0.
*** Bug 682765 has been marked as a duplicate of this bug. ***
I'll pull this over for F15 too.
Discussed at 2011-03-11 blocker review meeting. This issue is fixed and available for testing in anaconda-15.21-1. The issue has been accepted as a beta blocker.
anaconda-15.26-1.fc15 has been submitted as an update for Fedora 15. https://admin.fedoraproject.org/updates/anaconda-15.26-1.fc15
Package anaconda-15.26-1.fc15: * should fix your issue, * was pushed to the Fedora 15 testing repository, * should be available at your local mirror within two days. Update it with: # su -c 'yum update --enablerepo=updates-testing anaconda-15.26-1.fc15' as soon as you are able to. Please go to the following url: https://admin.fedoraproject.org/updates/anaconda-15.26-1.fc15 then log in and leave karma (feedback).
Tested anaconda 15.25 of F15-beta, issue still happened.
(In reply to comment #19) > Tested anaconda 15.25 of F15-beta, issue still happened. Hi Hurry, can you upload anaconda.log for this issue?
Created attachment 488984 [details] anaconda-logs.tar.gz
Discussed in the 2011-04-01 blocker review meeting. clumens will be look at this and try to fix.
I just tested anaconda-15.25-1 (F-15-Beta-TC1) using a kickstart that includes the 'reboot' command. The kickstart was delivered via http (ks=http://server/path/to/ks.cfg ). The installer does appear like it is rebooting, however it doesn't boot back up. It only shuts down. unmounting filesystems... /dev/pts done /selinux done /mnt/sysimage/boot done /mnt/sysimage/dev/pts done /mnt/sysimage/dev/shm done /mnt/sysimage/dev done /mnt/sysimage/proc/bus/usb done /mnt/sysimage/proc done /mnt/sysimage/selinux done /mnt/sysimage/sys done /mnt/sysimage done waiting for mdraid sets to become clean... sending termination signals...done sending kill signals...done rebooting system [ 426.833572] Restarting system. This was tested in a KVM guest on a F14 KVM host. On the same host, I confirmed that this is same behavior happens with a F-14 guest on my F-14 host. I'm not seeing any further problems here, it looks like it's behaving consistent with previous releases. I'll test on bare metal shortly, I have a feeling the failure to "reboot" is specific to virtualization. Hurry, can you confirm? Am I missing/forgetting something?
anaconda-15.26-1.fc15 has been pushed to the Fedora 15 stable repository. If problems still persist, please make note of it in this bug report.
(In reply to comment #23) > I just tested anaconda-15.25-1 (F-15-Beta-TC1) using a kickstart that includes > the 'reboot' command. The kickstart was delivered via http > (ks=http://server/path/to/ks.cfg ). The installer does appear like it is > rebooting, however it doesn't boot back up. It only shuts down. > > unmounting filesystems... > /dev/pts done > /selinux done > /mnt/sysimage/boot done > /mnt/sysimage/dev/pts done > /mnt/sysimage/dev/shm done > /mnt/sysimage/dev done > /mnt/sysimage/proc/bus/usb done > /mnt/sysimage/proc done > /mnt/sysimage/selinux done > /mnt/sysimage/sys done > /mnt/sysimage done > waiting for mdraid sets to become clean... > sending termination signals...done > sending kill signals...done > rebooting system > [ 426.833572] Restarting system. > This was tested in a KVM guest on a F14 KVM host. I can always reproduce it. > On the same host, I > confirmed that this is same behavior happens with a F-14 guest on my F-14 host. Do you mean it also happens when virt-installing F-14 guest on F-14 host? I can't reproduce it. > I'm not seeing any further problems here, it looks like it's behaving > consistent with previous releases. I'll test on bare metal shortly, I have a > feeling the failure to "reboot" is specific to virtualization. > Hurry, can you confirm? Am I missing/forgetting something? I used to test anaconda-15.25-1 (F-15-Beta-TC1) using a kickstart that doesn't include the 'reboot' command. After the kickstart auto-install, I manually clicked 'reboot', but after that it halted like: Running halt... disabling swap... /dev/sda3 unmounting filesystems... /dev/pts done /mnt/source done /selinux done /mnt/sysimage/boot done /mnt/sysimage/dev/pts done /mnt/sysimage/dev/shm done /mnt/sysimage/dev done /mnt/sysimage/proc/bus/usb done /mnt/sysimage/proc done /mnt/sysimage/selinux done /mnt/sysimage/sys done /mnt/sysimage done waiting for mdraid sets to become clean... sending termination signals...done sending kill signals...done halting system [ 426.833572] System halted. For me, this issue happens when I use kickstart to install F-15 both in kvm and bare metal.
(In reply to comment #24) > anaconda-15.26-1.fc15 has been pushed to the Fedora 15 stable repository. If > problems still persist, please make note of it in this bug report. I tested it and found that if the kickstart includes the 'reboot' command, anaconda can reboot as expected after installation. But if ks doesn't include 'reboot', anaconda still halts after manually clicking 'reboot' button. Reopen the bug.
(In reply to comment #26) > (In reply to comment #24) > > anaconda-15.26-1.fc15 has been pushed to the Fedora 15 stable repository. If > > problems still persist, please make note of it in this bug report. > > I tested it and found that if the kickstart includes the 'reboot' command, > anaconda can reboot as expected after installation. But if ks doesn't include > 'reboot', anaconda still halts after manually clicking 'reboot' button. Reopen > the bug. Thanks for the quick feedback Hurry, and good testing. I've also confirmed that doing a Kickstart install without 'reboot', and then clicking the Reboot button ... indeed halts the system (instead of rebooting it). I hesitate to file this as a new bug, since the current behavior is slightly different from the reported issue. I'll defer to how clumens would like to handle the issue. Also, based on your testing, I think the original use case is now sufficiently handled to cover the Beta release criteria. If a user includes 'reboot' in their kickstart, the installation will be unattended as expected and reboot properly into the installed system. My sense is that the new behavior you discovered no longer impacts the unattended install beta release criteria. [1] " Any installation method or process designed to run unattended must do so (there should be no prompts requiring user intervention) " I recommend we move this to F-15-Beta Nice-to-have, and/or file a new bug to track the remaining problem. [1] https://fedoraproject.org/wiki/Fedora_15_Beta_Release_Criteria
I'm going to take the lame way out on this bug. Given that we've met the release criteria (per comment #27) and are pretty far into edge case territory (comment #26), I say we close this for F15. For post-F15, I am actively working on getting rid of our own init and replacing it with systemd. As part of that, I'm drastically simplifying the process management and rebooting code. Hopefully, I'll have this fixed there. I'll make sure to put this case on the list of things to check.
*** Bug 695445 has been marked as a duplicate of this bug. ***
(In reply to comment #27) [...] > I've also confirmed > that doing a Kickstart install without 'reboot', and then clicking the Reboot > button ... indeed halts the system (instead of rebooting it). [...] I can confirm this behaviour in F15 Beta. While not ideal, I can live with it, provided it gets fixed in F16 and does not bite me in RHEL7 once that materialises. (I control the kickstart files, so adding reboot statement is easy, even though for some tests I prefer to have the system actually not reboot immediately after install) Chris; what about tracking this for F16, do you want a cloned bug of this or shall this one be re-opened and assigned to release F16?
I no longer see this bug in F16 Beta. If I have 'reboot' in the ks file the system reboots as expected. If I do not have 'reboot' in the ks file then it reboots just fine after clicking OK in the final dialog box anaconda shows. Thanks for the fix.