Instance successfully booted with above noted workaround. Created a ingress security group rule to allow SSH access to the instance. The rule applied as expected and ssh access was allowed. Created an egress security group rule to allow HTTP access out from the instance. The rule applied as expected all traffic out of the VM was blocked excluding http. Created a new volume, successfully attached it to the booted instance.
Added a new ISO. ISO was successfully downloaded and installed. Was successfully able to download the ISO from secondary storage via the UI. I was unable to delete the ISO using the UI. (No button present) I created a new project, the UI was inconsistent http://ianduffy.ie/cloudstack/testing45/project-ui.png I noticed the baremetal-system-account as documented over here: https://cwiki.apache.org/confluence/display/CLOUDSTACK/Baremetal+Advanced+Networking+Support It felt confusing/unclear. It should be hidden like the system account in my view. Clicking add guest network and then pressing cancel resulted in an api error: http://ianduffy.ie/cloudstack/testing45/Screen%20Shot%202014-10-02%20at%2002.12.11.png Successfully created a new account and created new instances with this account. I was able to submit an invalid email for this e.g. not <something>@<something>.<something> Successfully created a subdomain, UI didn't display it until the parent domain was collapsed and expanded. Creating enough nested subdomains allowed the tree to go off the screen. no scroll bar was provided. http://ianduffy.ie/cloudstack/testing45/Screen%20Shot%202014-10-02%20at%2002.16.48.png If I: 1) Create a subdomain of root labelled test 2) Create a subdomain of the subdomain created in step one, again labelling it test e.g. test/test 3) I rename test subdomain directly under root to rubbishdata 4) I attempt to rename rubbishdata back to test I get a message saying: "specified domain id with name 'test' already exists in the system" On 2 October 2014 01:55, Ian Duffy <i...@ianduffy.ie> wrote: > Appear to be able to work around the virtual router issue by modifying the > version field in the domain_router table. It will read Cloudstack Release > Mon Sep 29 23:17:27 UTC 2014 changing to Cloudstack Release 4.5 Mon Sep 29 > 23:17:27 UTC 2014 allowed me to boot an instance (note the 4.5 before Mon). > > will continue testing.... > > On 2 October 2014 01:51, Ian Duffy <i...@ianduffy.ie> wrote: > >> Hi All, >> >> Using the following environment: >> >> - XenServer 6.2 >> - The RPM build that was created on October 1st 2015 4:51am >> http://jenkins.buildacloud.org/job/package-rhel63-master/3361/ >> - The system vm that was build on September 29th 2015 7:07am >> http://jenkins.buildacloud.org/job/build-systemvm64-master/425/ >> >> I brought up a basic zone using marvin to do all the configuration. If >> anybody wants re-create the exact same environment there is code over at >> https://github.com/imduffy15/cookbook_cloudstack_wrapper adjust >> chef_configuration.json as necessary to change the repo and systemvm and >> then do a vagrant up (if your on windows expect issues). >> >> Configuration was successful, my systemvms were successfully created: >> http://ianduffy.ie/cloudstack/testing45/version%2bsystemvms-up.png >> >> There was an issue with a table header on the system vms page: >> http://ianduffy.ie/cloudstack/testing45/strange-labeling-on-ui.png >> >> Another labelling issue was noted on the alerts section on the initial >> screen: >> http://ianduffy.ie/cloudstack/testing45/Screen%20Shot%202014-10-02%20at%2001.40.18.png >> >> The cluster configuration table under infrastructure had a white block as >> its header, on hovering text would appear: >> http://ianduffy.ie/cloudstack/testing45/cluster-table-renders-oddly.png >> >> The system vms complained about a file not being found: >> http://ianduffy.ie/cloudstack/testing45/system-vm-reports-script-not-found.png >> >> Rapidly clicking the search button resulted in duplicate entries: >> http://ianduffy.ie/cloudstack/testing45/clicking-the-search-button-multiple-times-results-in-duplicates.png >> >> Add instance UI had a random scroll bar for no reason: >> http://ianduffy.ie/cloudstack/testing45/add-instance-scroll-bar-for-no-reason.png >> >> I attempted to create an instance, it failed. I believe this is due to an >> issue with the version field populated for the virtual router logs >> available at https://gist.github.com/imduffy15/5879718ce38f8a67bebd >> After this error was created I got random prompt boxes on the UI: >> http://ianduffy.ie/cloudstack/testing45/Screen%20Shot%202014-10-02%20at%2001.40.18.png >> >> I stopped my testing at this stage. >> >> Thanks, >> >> Ian >> >> On 1 October 2014 17:26, Daan Hoogland <daan.hoogl...@gmail.com> wrote: >> >>> I have been looking at using my collegues integration work for the master >>> branch. So far the packaging part is holding my back as it uses >>> 'SNAPSHOT' >>> in the build and does not allow for patch levels (in package.sh) I am >>> looking at a solution. >>> >>> what I am thinking of is making rpms like >>> >>> >>> cloudstack-<package>-<version-including-SNAPSHOT>-<version>-<buildnumber-from-jenkins>.el6_64.rpm >>> >>> this seems quite verbose but it gives us space to do continuous >>> integration >>> on the artifact. I am sure a scheme like this can be deviced for .deb >>> packages as well. >>> >>> thoughts? >>> >>> The actual testing, for as little far as I got to do, did not yield >>> issues >>> yet. >>> >>> Daan >>> >>> On Wed, Oct 1, 2014 at 5:28 PM, Marcus <shadow...@gmail.com> wrote: >>> >>> > I'd be curious to know if it works by re-adding the host. I remember >>> > running into situations like this in the past, where the mgmt server >>> was >>> > either not fully up or rejecting agents for some reason. >>> > >>> > On Wed, Oct 1, 2014 at 7:01 AM, Wido den Hollander <w...@widodh.nl> >>> wrote: >>> > >>> > > >>> > > >>> > > On 09/30/2014 09:26 AM, Daan Hoogland wrote: >>> > > > H all, >>> > > > >>> > > > As you might remember, we agreed on a test day for the master >>> branch on >>> > > > Wednesday the 1st October, during our last meeting in Schihol-Rijk >>> with >>> > > > some folks. Please feel invited to test master so that we can >>> report to >>> > > our >>> > > > RM for 4.5 for the current state and he can have a short and quick >>> > > journey >>> > > > to releasing. >>> > > > >>> > > >>> > > I didn't forget! Just tested by upgrading from 4.4.1 (RC) to 4.5.0 >>> and >>> > > that worked, but my KVM agent couldn't connect anymore afterwards: >>> > > >>> > > 2014-10-01 14:59:08,789 INFO [utils.nio.NioClient] >>> > > (Agent-Selector:null) Connecting to cloudstack.XXXX:8250 >>> > > 2014-10-01 14:59:08,809 ERROR [utils.nio.NioConnection] >>> > > (Agent-Selector:null) Unable to initialize the threads. >>> > > java.io.IOException: Connection closed with -1 on reading size. >>> > > at com.cloud.utils.nio.Link.doHandshake(Link.java:509) >>> > > at com.cloud.utils.nio.NioClient.init(NioClient.java:78) >>> > > at >>> com.cloud.utils.nio.NioConnection.run(NioConnection.java:111) >>> > > at java.lang.Thread.run(Thread.java:744) >>> > > >>> > > Any ideas? When I try to telnet to 8250 the connection is closed >>> > > immediately by the mgmt server. Nothing in the logs. >>> > > >>> > > So this is currently preventing me from testing the KVM agents. >>> > > >>> > > Wido >>> > > >>> > > >>> > > > optimistically yours, >>> > > > >>> > > >>> > >>> >>> >>> >>> -- >>> Daan >>> >> >> >
