On 07/17/10 - 06:07:00PM, Matthias Bolte wrote: > For example this one. I tested it with and without this patch. > > Without this patch we get > > virsh # snapshot-create test1 this-file-does-not-exist.xml > error: Failed to open file 'this-file-does-not-exist.xml': No such > file or directory > virsh # > > With this patch applied we get > > virsh # snapshot-create test1 this-file-does-not-exist.xml > virsh # > > Making it look like it succeeded. I assume it's the same pattern in > all places where you removed the call to virshReportError on > virFileReadAll failure. > > Ah, okay. Now I see that you fixed this in patch 6/8. Now the error is > more verbose, but that's okay I think. > > virsh # snapshot-create test1 this-file-does-not-exist.xml > error: Failed to read contents of 'this-file-does-not-exist.xml' > error: Failed to open file 'this-file-does-not-exist.xml': No such > file or directory > virsh # > > Maybe you should merge patch 4 and 6 into one. That way error > reporting for virFileReadAll isn't broken in between those two > commits. Yep, you hit it exactly. I was trying to get the error reporting to be more explicit, but I did it in two stages. Merging this and 6 is not a bad idea, I'll do that when I commit. > > ACK. Thanks, -- Chris Lalancette -- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list