Bug 1152485
| Summary: | Anaconda with root=live:http boot option fetches squashfs.img for each NIC brought up in initramfs | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 7 | Reporter: | Radek Vykydal <rvykydal> |
| Component: | dracut | Assignee: | dracut-maint |
| Status: | CLOSED ERRATA | QA Contact: | Release Test Team <release-test-team-automation> |
| Severity: | unspecified | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 7.0 | CC: | dracut-maint-list, harald, mbanas, mkolman, mkovarik, mnavrati, rvykydal, tlavigne |
| Target Milestone: | rc | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: |
Due to a bug, the initramfs exited after a successful download of a live image, and the download of the live image was retried for every network interface. Now, the script only exits on error, and the live image is downloaded only once.
|
Story Points: | --- |
| Clone Of: | Environment: | ||
| Last Closed: | 2015-11-19 12:51:06 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: | |
| Embargoed: | |||
|
Description
Radek Vykydal
2014-10-14 09:21:25 UTC
anaconda-netroot.sh is looking for /dev/root as a lock, but I'm guessing that we end up in a race between several nics running it. We're going to need a good way to lock things, and possibly retry if, for example, the first one can't get to whatever network the squashfs.img is on. (In reply to bcl from comment #2) > anaconda-netroot.sh is looking for /dev/root as a lock, but I'm guessing > that we end up in a race between several nics running it. We're going to > need a good way to lock things, and possibly retry if, for example, the > first one can't get to whatever network the squashfs.img is on. This is a possible design we came up with when braistorming this with Radek a bit yesterday: For each download request "thread": 0. check for the existence of a "DONE" file, exit if it exists 1. check for the existence of a "IN_PROGRESS" file -> if it exist busy wait with sleep as long as it is there -> if it does not exist, create it 2. start the download 3. once the download finishes -> if successful delete the "IN_PROGRESS" file, write the "DONE" file and exit -> if the download failed delete the "IN_PROGRESS" file and exit It happens only when root=live:http option is used to fetch squashfs.img So, as Martin found out it works ok with inst.stage2=, ie the squashfs.img is downloaded only once. This is an issue eg for pxe boot to live Atomic. Comment #10 is a suggested patch to modules.d/90livenet/livenetroot.sh from dracut. ouch.. misread the comment. sorry dracut-033-335.el7 Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://rhn.redhat.com/errata/RHBA-2015-2091.html |