| Summary: | dracut is unable to boot while doing BFO install, drops to dracut shell | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | stan <gryt2> | ||||||||
| Component: | distribution | Assignee: | Václav Pavlín <vpavlin> | ||||||||
| Status: | CLOSED CURRENTRELEASE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||||||
| Severity: | unspecified | Docs Contact: | |||||||||
| Priority: | unspecified | ||||||||||
| Version: | rawhide | CC: | dennis, dracut-maint-list, g.kaviyarasu, gryt2, harald, jonathan, kevin, mkolman, vanmeeuwen+fedora, zbyszek | ||||||||
| Target Milestone: | --- | ||||||||||
| Target Release: | --- | ||||||||||
| Hardware: | Unspecified | ||||||||||
| OS: | Unspecified | ||||||||||
| Whiteboard: | |||||||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||||||
| Doc Text: | Story Points: | --- | |||||||||
| Clone Of: | Environment: | ||||||||||
| Last Closed: | 2016-02-15 15:04:03 UTC | Type: | Bug | ||||||||
| Regression: | --- | Mount Type: | --- | ||||||||
| Documentation: | --- | CRM: | |||||||||
| Verified Versions: | Category: | --- | |||||||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||||||
| Attachments: |
|
||||||||||
kernel command line: repo=http://download.fedoraproject.org/pub/fedora/linux/development/rawhide/x86_64/os/ initrd=http://download.fedoraproject.org/pub/fedora/linux/development/rawhide/x86_64/os/images/pxeboot/initrd.img BOOT_IMAGE=http://download.fedoraproject.org/pub/fedora/linux/development/rawhide/x86_64/os/images/pxeboot/vmlinuz What is BFO? reassigning to anaconda. I think that in this case BFO means: https://boot.fedoraproject.org/ Looks like the rd.timeout in the boot.fedoraproject.org needs to be increased. Well, there's currently no timeout set... what would you recommend we set it to? I notice that when the downloads take about 5 minutes, this error doesn't occur. When they are slow, taking several hours, it usually occurs. The speed ratio between the two is about 50, so I suggest a 60 second timeout, which gives a little room for error. Done. Please test and let me know if there's still any issue... Thanks. I'll try it tonight. Created attachment 1127658 [details]
File that dracut generated on failure
Created attachment 1127659 [details]
journalctl -r output from failure
The same error occurred, but when I looked in journalctl I saw that the download had failed. I watched if for a few minutes when I started it, and it was running at about 66KB/sec. But it stopped shortly after I left it unattended. It might have been because of maintenance. Feb 16 05:57:10 localhost dracut-initqueue[696]: Warning: /dev/root does not exist Feb 16 05:57:10 localhost systemd[1]: Received SIGRTMIN+20 from PID 700 (plymouthd). Feb 16 05:57:09 localhost dracut-initqueue[696]: Warning: Could not boot. losetup: /tmp/curl_fetch_url1/install.img: Warning: file does not fit into a 512-byte sector; the end of the file will be ignored. Feb 16 05:54:05 localhost kernel: loop: module loaded Feb 16 05:54:03 localhost dracut-initqueue[696]: curl: (18) transfer closed with 379650412 bytes remaining to read Feb 16 05:54:03 localhost dracut-initqueue[696]: [1.9K blob data] Feb 16 05:53:40 localhost dracut-initqueue[696]: [1.9K blob data] Feb 16 05:53:14 localhost dracut-initqueue[696]: [2.0K blob data] Feb 16 05:52:47 localhost dracut-initqueue[696]: [1.9K blob data] Feb 16 05:52:21 localhost dracut-initqueue[696]: [2.0K blob data] Feb 16 05:51:55 localhost dracut-initqueue[696]: [2.0K blob data] Feb 16 05:51:29 localhost dracut-initqueue[696]: [2.0K blob data] Feb 16 05:51:03 localhost dracut-initqueue[696]: [2.0K blob data] Feb 16 05:50:37 localhost dracut-initqueue[696]: [2.0K blob data] Feb 16 05:50:11 localhost dracut-initqueue[696]: [237B blob data] Feb 16 05:50:11 localhost dracut-initqueue[696]: Dload Upload Total Spent Left Speed Feb 16 05:50:11 localhost dracut-initqueue[696]: % Total % Received % Xferd Average Speed Time Time Time Current Feb 16 05:50:11 localhost dracut-initqueue[696]: Warning: can't find installer mainimage path in .treeinfo Feb 16 05:50:10 localhost dracut-initqueue[696]: anaconda fetching installer from http://download.fedoraproject.org/pub/fedora/linux/development/rawhide/x86_64/os/ Feb 16 05:50:09 localhost dracut-initqueue[696]: RTNETLINK answers: File exists I ran this morning, and even though there were a couple of lulls with no transmission, the download completed, and I was able to get into anaconda. Where I had an error I've been getting and reported it in another bugzilla. The download speed was about 138kB/sec, so took about 45 minutes total. That seems like an adequate test. So, this is probably fixed, but I'll continue to monitor it as I try to complete a rawhide installation using BFO. |
Created attachment 1122895 [details] This is the report that dracut generated when the error occurred. Description of problem: While using BFO to do an install of rawhide, the system gets to the point where it says basic system installed. It then goes into dracut timeouts, and drops to a dracut shell. Version-Release number of selected component (if applicable): dracut-044-17.git20160108.fc24 How reproducible: Happens about half the time when doing a BFO install. Seems to happen every time with slow download of OS, almost never when download is fast. Steps to Reproduce: 1. Do a BFO install of rawhide x86_64 2. If downloads are slow, the usual case, pretty good chance this error will happen 3. Actual results: See this error Expected results: Proceed to download of stage2 and anaconda install Additional info: The information that comes up at the dracut shell prompt when help is typed is misleading. It would be better to do an ls of the /bin directory and tell the user that those are the commands available.