On 03/20/2012 03:29 PM, Eric Blake wrote:
The oVirt developers have stated that the real reasons they want to have qemu reuse existing volumes when creating a snapshot are: 1. the management framework is set up so that creation has to be done from a central node for proper resource tracking, and having libvirt and/or qemu create things violates the framework, and 2. qemu defaults to creating snapshots with an absolute path to the backing file, but oVirt wants to manage a backing chain that uses just relative names, to allow for easier migration of a chain across storage locations.
/* create the stub file and set selinux labels; manipulate disk in * place, in a way that can be reverted on failure. */ - fd = qemuOpenFile(driver, source, O_WRONLY | O_TRUNC | O_CREAT, + fd = qemuOpenFile(driver, source, + O_WRONLY | (reuse ? 0 : O_TRUNC | O_CREAT), &need_unlink, NULL); if (fd < 0) goto cleanup;
I just realized this is a pointless open in the reuse case, since the very next statement closes the fd (in other words, we are using open for the side-effect of creation, but reuse says no creation is necessary). I plan on squashing this: diff --git i/src/qemu/qemu_driver.c w/src/qemu/qemu_driver.c index ca81e75..4286b2a 100644 --- i/src/qemu/qemu_driver.c +++ w/src/qemu/qemu_driver.c @@ -9884,12 +9884,13 @@ qemuDomainSnapshotCreateSingleDiskActive(struct qemud_driver *driver, /* create the stub file and set selinux labels; manipulate disk in * place, in a way that can be reverted on failure. */ - fd = qemuOpenFile(driver, source, - O_WRONLY | (reuse ? 0 : O_TRUNC | O_CREAT), - &need_unlink, NULL); - if (fd < 0) - goto cleanup; - VIR_FORCE_CLOSE(fd); + if (!reuse) { + fd = qemuOpenFile(driver, source, O_WRONLY | O_TRUNC | O_CREAT, + &need_unlink, NULL); + if (fd < 0) + goto cleanup; + VIR_FORCE_CLOSE(fd); + } origsrc = disk->src; disk->src = source; -- Eric Blake eblake@redhat.com +1-919-301-3266 Libvirt virtualization library http://libvirt.org