Leo, great work thanks
On Wed, Jun 11, 2014 at 11:39 PM, Leo Simons <lsim...@schubergphilis.com> wrote: > Hey folks, > > On 6/11/14, 6:36 PM, "Mike Tutkowski" <mike.tutkow...@solidfire.com> wrote: >>Yeah, I am concerned about 4.4 getting farther behind schedule, as well, >>but I agree with David that we should not cancel it. >> >>I know it might be a pain, but I wonder if the RM would be willing to >>"nag" >>people every few days (just an e-mail to dev@) about the current list of >>blockers and their progress and to see if people need help and others >>might >>be willing and able to do so. > > Funny you should mention that. I¹m not the RM obviously, but Daan asked me > if I could have a fresh look at the list of blockers and criticals. If you > want, consider me a temporary junior administrative assistant to the > release manager ;-). What¹s a chore for Daan is a good way for me to learn > more cloudstack! > > Here¹s this afternoon's blockers against 4.4 ordered by creation date: > > * Bug CLOUDSTACK-6602 > [UI] createNetworkACL API action param value passed incorrectly > Assignee: Jessica Wang > Created: 5/8/14 5:55 > Updated: 6/3/14 13:08 > Leo¹s evaluation: seems like a blocker; > _probably_ simple fix if you know the javascript? > Leo¹s suggestion: fix for 4.4 > > > > * Bug CLOUDSTACK-6779 > [OVS] Expunging VM (deleting vif) deletes all the rules from ovs > bridge flow table > Assignee: Murali Reddy > Created: 5/27/14 12:49 > Updated: 6/6/14 12:43 > Leo¹s evaluation: seems like a blocker, work is in progress > Leo¹s suggestion: fix for 4.4 > > > > * Bug CLOUDSTACK-6791 > [Automation] DeleteNetworkCmd fails with NullPointerException > Assignee: Santhosh Kumar Edukulla > Created: 5/28/14 1:01 > Updated: 6/8/14 21:28 > Leo¹s evaluation: stalled on info from tester > Leo¹s suggestion: downgrade to major > > > > * Test CLOUDSTACK-6861 > [Automation] Test case test_01_snapshot_root_disk during mount > Assignee: **Unassigned** > Created: 6/9/14 3:50 > Updated: 6/9/14 3:55 > Leo¹s evaluation: looks like it could be a test environment issue > instead of a code issue? > Leo¹s suggestion: downgrade to major > > > > * Test CLOUDSTACK-6863 > [Automation] test_09_delete_detached_volume failing inconsistently with > during simulator run > Assignee: **Unassigned** > Created: 6/9/14 4:03 > Updated: 6/9/14 4:04 > Leo¹s evaluation: seems like a blocker, looks scary > Leo¹s suggestion: fix for 4.4, mark as bug, not test > > > > * Test CLOUDSTACK-6862 > [Automation] Test case > TestDeployvGPUenabledVM.test_deploy_vgpu_enabled_vm > faling during BVT > Assignee: **Unassigned** > Created: 6/9/14 3:55 > Updated: 6/10/14 21:01 > Leo¹s evaluation: test for unfinished new feature is failing > Leo¹s suggestion: if not easy to fix, defer feature to 4.5 > > I was going to give my suggestions to Daan to try and make a call on, but I > guess it¹s better if that can be done by y¹all, right here. > > Note a lot of these are very recent coming out of ongoing QA, so I imagine > this list is still to grow before it shrinks. > > Next, here¹s a bunch of older critical issues against 4.4: > > * CLOUDSTACK-4593 > [VMWARE] [Upgrade]Livestorage Migration & VM Snapshot features are not > fully functional after upgrade > Sateesh Chodapuneedi > 9/3/13 6:26 > 12/6/13 10:27 > nasty bug since 4.2 was not fixed in 4.3 > defer to 4.5 > > * CLOUDSTACK-4247 > [VMWARE]Network read/write statistics is zero always > Likitha Shetty > 8/12/13 10:00 > 12/10/13 4:04 > bug since 4.1, prio went down/up few times > defer to 4.5 > > * CLOUDSTACK-5356 > Xenserver - Failed to create snapshot when secondary store was made > unavaibale for about 1.5 hour leaving behind snapshot in " > CreatedOnPrimary" state. The subsequent scheduled snapshot also > failed > edison su > 12/4/13 0:48 > 12/20/13 0:21 > nasty bug on 4.3 was bumped to 4.4 (in xenserver?) > defer to 4.5 > > * CLOUDSTACK-5357 > Xenserver - Failed to create snapshot due to "unable to destroy > task(com.xe nsource.xenapi.Task@67d312d6) on > host(23af93a0-93ff-40cb-ba11-a11d1b884d37)" when secondary store was > unavaiable for 1 and 1/2 hours and then brought up.. > edison su > 12/4/13 0:57 > 12/20/13 0:22 > similar to 5356 > defer to 4.5 > > * CLOUDSTACK-5469 > Snapshot creation fails with following exception - "Failed to backup > snapshot: qemu-img: Could not delete snapshot > '89eced14-9121-44a7-bb97-26b567795726': -2 (No such file or directory)" > edison su > 12/12/13 0:13 > 12/20/13 0:23 > similar to 5356 > defer to 4.5 > > * CLOUDSTACK-4492 > [object_store_ref] Attaching volume to a vm is failing after upgrade > if the volume was uploaded before upgrade > edison su > 8/26/13 7:04 > 12/31/13 6:54 > bug in upgrade from 3 to 4 > downgrade to major > > * CLOUDSTACK-5005 > issue with stopping vms parallelly > Kelven Yang > 10/30/13 12:14 > 1/9/14 21:55 > vmsync, might be fixed?, needs retesting > test if fixed in 4.4, bump to 4.5 if not > > * CLOUDSTACK-5736 > KVM - Recurring Snapshots - Snapshots reported as being in "BackedUp" > state even though a Nullpointer exception is seen during post processing > of snapshot. > Kelven Yang > 1/3/14 2:21 > 2/3/14 23:05 > vmsync, might be fixed?, needs retesting > test if fixed in 4.4, bump to 4.5 if not > > * CLOUDSTACK-5738 > Recurring Snapshots - Few Snapshots are stuck in "Creating" state for > more than 30 minutes. > Kelven Yang > 1/3/14 2:42 > 2/3/14 23:06 > vmsync, might be fixed?, needs retesting > test if fixed in 4.4, bump to 4.5 if not > > * CLOUDSTACK-6051 > VR Rolling upgrade: Make the numbers of Routers parallely being > upgraded as configurable > Kishan Kavala > 2/7/14 5:12 > 2/7/14 5:12 > don't understand, needs description > evaluate > > * CLOUDSTACK-6063 > CLONE - Non windows instances are created on XenServer with a vcpu-max > above supported xenserver limits > Harikrishna Patnala > 2/7/14 21:42 > 2/7/14 21:46 > hack/workaround in 4.3, check if fixed in 4.4 > evaluate if fixed > > * CLOUDSTACK-5899 > Contrail:MS: Exceptions in MS logs on a fresh install, syncDomain > java.lang.NullPointerException > Suresh Balineni > 1/17/14 22:41 > 4/5/14 0:14 > eek, NPE! Was not fixed in 4.3 > defer to 4.5 > > * CLOUDSTACK-4912 > API docs are missing some APIs > David Nalley > 10/21/13 18:10 > 4/5/14 0:14 > docs/jenkins issue > downgrade to major, remove version > > * CLOUDSTACK-5358 > API: synchronization on the object is broken > Kelven Yang > 12/4/13 1:11 > 4/5/14 0:14 > vmsync, might be fixed? Needs eval > evaluate > > * CLOUDSTACK-5746 > [HyperV]Can't access vm console from IE browser > Devdeep Singh > 1/3/14 6:47 > 4/22/14 4:56 > bug for IE < 11, IE is unsupported > downgrade to major, defer to 4.4 > > * CLOUDSTACK-6177 > CS does XS master switch, which may cause weid XS behavior > Anthony Xu > 2/26/14 23:23 > 4/22/14 18:33 > eek, failure cascade! Needs attention? > defer to 4.5 > > * CLOUDSTACK-5012 > Bad data inserted into physical network labels for Zone Create Wizard > using VMWare > Sateesh Chodapuneedi > 10/31/13 4:10 > 4/28/14 10:21 > complex issue, needs test automation > defer to 4.5 > > * CLOUDSTACK-5806 > Storage types other than NFS/VMFS can't overprovision > Unassigned > 1/6/14 19:43 > 4/30/14 17:35 > needs attention > defer to 4.5 > > * CLOUDSTACK-5212 > Need Support for the LXC for the Report sockets CS-4908 > Kishan Kavala > 11/20/13 6:46 > 5/9/14 19:25 > incomplete new feature > downgrade to major, defer to 4.5 > > * CLOUDSTACK-5372 > Xenserver - SR not being recreated when the Primary storage is brought > down and brought back up again resulting in not being able to start > the > Vms that have their volumes in this primary store. > Anthony Xu > 12/4/13 22:53 > 5/9/14 19:25 > failure cascade! > defer to 4.5 > > As you can see, as far as I understand (which is not very farŠ), most of > these older issues are definitely serious. In other words, I think most of > them definitely deserve to be considered critical bugs, they¹re not > particularly mis-triaged. But, they¹re very specific / difficult to > reproduce / limited to a specific combination of components, so they > should perhaps not hold up a release, given where cloudstack is at right > now. For many of them, that¹s the same choice that was made when baking > the 4.3 release. It¹d obviously be great to not have to bump them again, > but, given this release is so late, it seems a reasonable choice. > > Then there¹s another bucket of issues related to something called ³vmsync² > which need some follow up from someone that understands that stuff. > > CLOUDSTACK-6063 also deserves a highlight. AIUI this was a dirty/quick fix > for 4.3 that should have received a proper fix in then-master now-4.4. I > didn¹t find yet whether it did. > > I think tomorrow I¹m digging through the criticals that have seen updates > after may 1st, of which there are currently **68**. Comparing bug creation > rate with bug resolution rate, there will never be a 4.4 release if all > those have to be fixedŠso I imagine the only available approach is to > defer a lot of _those_ to 4.5, too :-/. > > > Hope this helps! Cheers, > > > Leo > -- Daan
