On Fri, 2013-12-13 at 14:13 -0800, Adam Williamson wrote: > I meant to write this test case anyhow, but today's brown paper bag bug > - https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=1040669 - gives > it a certain sense of importance, so I thought I'd best get it done now. > > I wrote > https://fedoraproject.org/wiki/QA:Testcase_anaconda_lvmthinp_rootfs , > and I propose we add it to the installation validation matrix for F21 > and later, with release level Beta, as it matches this criterion: > > https://fedoraproject.org/wiki/Fedora_20_Beta_Release_Criteria#Guided_partitioning > > "When using the guided partitioning flow, the installer must be able > to: ... Complete an installation using any combination of disk > configuration options it allows the user to select" > > LVM thinp is now one of the options on the 'filesystem type' dropdown on > the Installation Options screen, which is what 'disk configuration > options' in that criterion is talking about. > > Anyone have feedback on the test case, or the proposed inclusion in the > matrices? I based the test case on the ext4 rootfs one but tidied it up > somewhat, and placed stronger restrictions on changing other > configuration during the test; I will probably go through all the > filesystem-y test cases and propose similar changes, if we think that's > a good idea. We should try to make the whole set consistent. I went ahead and stuck the thinp test case in the F21 template, for now, since no-one raised any issues with it. Beyond that, though, even a *small* rework of the existing partitioning test cases exposes a bit of a problem. The partitioning test cases we do have are basically still oldUI holdovers, they were written explicitly to the old anaconda UI. There wasn't so much filesystem selection in autopart in oldUI, so the test cases are split between just exercising the different autopart algorithms oldUI had (use all space, use free space, wipe Linux partitions only) and a few desultory tests of different filesystems in custom partitioning. oldUI did let you pick whether to use LVM or not, but the old test cases just sort of conveniently ignored that distinction. newUI has filesystem options in guided partitioning, which makes things a bit more complex. right now we just have the old autopart test cases slightly tweaked for newUI's guided partitioning, and the old filesystem test cases not really accounting for the fact that you can do filesystem selection in guided partitioning. My off-the-top-of-my-head idea for revising it would be to do the kind of 'matrix judo' I've been fond of recently, with small matrices with different column headings according to what's being exercised: further modify the 'autopart' tests to properly cover the various possibilities in 'guided partitioning' for newUI, then have a matrix with those tests as the rows and the column headings being 'LVM, standard partition, btrfs, lvm thinp'. Actually - it'd basically just be the 'guided installation' table from https://fedoraproject.org/wiki/User:Adamwill/Draft_storage_matrix#Guided_installation without all the other bits, basically. Do we think it's worth taking that bit out of the larger storage rework proposal and sticking it in the current matrix, replacing the appropriate existing test cases? It would be an hour or two's work for me, I guess. Eh, I guess I'll just draft it up and send it out and see what you all think. -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net -- test mailing list test@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe: https://admin.fedoraproject.org/mailman/listinfo/test
