Bug 508258
| Summary: | qemu init script does nothing useful by default | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Pilar Bravo <pbravo> |
| Component: | qemu | Assignee: | Glauber Costa <gcosta> |
| Status: | CLOSED NOTABUG | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | medium | Docs Contact: | |
| Priority: | low | ||
| Version: | rawhide | CC: | dwmw2, gcosta, itamar, jaswinder, markmc, virt-maint |
| Target Milestone: | --- | Keywords: | Reopened |
| Target Release: | --- | ||
| Hardware: | i686 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2009-08-07 11:08:29 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
| Bug Depends On: | |||
| Bug Blocks: | 498969 | ||
|
Description
Pilar Bravo
2009-06-26 10:59:34 UTC
To the best of my knowledge, this is not a bug. The init script search for a directory containing the runtime environment to the target architecture (gnemul). The qemu package does not provide it, so it comes as no surprise. If you download gnemul and put it into /usr, it should work. If there is disagreement on this, please reopen Let there be disagreement!
It's actually looking for e.g. /usr/qemu-${machine}, but that's a minor detail
Where does one download this runtime environment?
IMHO, we shouldn't be shipping an initscript if it doesn't do anything useful by default
Maybe the initscript should register the binary handler and let a missing /usr/qemu-${machine} be a runtime error? e.g. if the initscript runs at boot and you later download the runtime
Putting it on our target list for F-12
yeah, our script in fact looks for /usr/qemu-${machine}.
I have no idea where to get it, since I don't use linux-user for years now.
Maybe people actually using can tell us about it?
Anyway, what I do know is that we have no way at all to ship it in Fedora
About your suggestion, it does nothing but postpone the error. And the way it works now, the user at least isn't misguided, since the binary handler is not present, so I slightly prefer the way it is.
But I'm fine with either approach.
I'm dubious about the usefulness of this feature, but you're right - making it a runtime error doesn't help much. Closing this as NOTABUG - the binary handlers will only work if there is a qemu-user runtime present Where to get such a runtime is another question. We don't know where :-) It would be nice to broadly hear user's request if possible, but from my side, I am 100 % okay with just removing it. I don't see any benefit it brings |