On Mon, Nov 04, 2019 at 10:06:40AM +0100, Stefan Hajnoczi wrote:
On Fri, Nov 1, 2019 at 1:18 PM Ján Tomko <jtomko(a)redhat.com>
wrote:
> + if (qemuSecuritySetDaemonSocketLabel(driver->securityManager, vm->def)
< 0)
> + goto cleanup;
> + fd = qemuOpenChrChardevUNIXSocket(chrdev);
> + if (fd < 0)
> + goto cleanup;
> + if (qemuSecurityClearSocketLabel(driver->securityManager, vm->def) <
0)
> + goto cleanup;
qemuSecurityClearSocketLabel() is not called in the
qemuOpenChrChardevUNIXSocket() error code path. Is this correct?
> +static void
> +qemuExtVirtioFSdStop(virQEMUDriverPtr driver,
> + virDomainObjPtr vm,
> + virDomainFSDefPtr fs)
> +{
The daemon stops automatically when the vhost-user socket is closed by
QEMU. Is it necessary to implement an explicit stop function?
That's good, but we've generally wanted to be explicit about cleaning
things up to cope with unexpected circumstances. In particular QEMU
can get itself stuck as a zombie if there's a dead disk, so it is
worth tearing down virtiofsd explicitly.
Regards,
Daniel
--
|:
https://berrange.com -o-
https://www.flickr.com/photos/dberrange :|
|:
https://libvirt.org -o-
https://fstop138.berrange.com :|
|:
https://entangle-photo.org -o-
https://www.instagram.com/dberrange :|