-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 While testing this, I ran it against parted without the refactor patch, expecting it to fail. I wa surprised to see that it passed. On closer investigation, it seems the existing dm code has a bug. When it failed to remove an in use partition, it just threw a warning exception where the only option was IGNORE, and carried on as if it was a success, even if the exception went unhandled. This made parted /dev/sde rm 1 exit with a success status even though the partition was in use.
Now that I think about it ( and test manualy ), the busy partition warning should have been thrown and caused an error exit before actually trying to remove the partition. I'm not sure why this isn't happening when I run the test suite. On 01/11/2012 09:29 PM, Phillip Susi wrote: > From: Phillip Susi <[email protected]> > > New test to verify that this commit works: > libparted: refactor device-mapper partition sync code > --- -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQEcBAEBAgAGBQJPD22pAAoJEJrBOlT6nu75UTEIANR+B/M27f089uhS9Ka1adXD +U2Q+JNRF9w63lJYZKd8EYxHUCB7KV2GO9fyo3z2NgUIbryVRMikscqtRlIhDyvp Y8wnyI5xnojYRXDyi+eZNXiuSAv4eudDiAHnvP2p3UvJQBP+NGJA7zQSTwtzCGZU In5ObHXaAw4kqLC4UGwnh7ww6o/J2Ys9Wwo1wHSM6PNktKz9QfrnmVMn6+l1TVRK q4NlFhiOEX26T/HKKN0nirE+9iT+lBXvDnkkAPXK5aAbXEOyy43ySp1EkUdSzH2s xP12Afjl1eSPdAsYWvadh40p/lVM0wmcnIzV2LKbZg4UoS/wfdG/Lhms7XE0hVY= =iLGx -----END PGP SIGNATURE-----
