How does HA work?
Hello, Since I'm not a programmer reading the code doesn't make too much sense to me. Can anyone explain to me how HA works (for VMs expecially)? Who monitors these HA VMs? Can I run multiple monitors, for HA? Any recommendations, best practices, gotchas? Thanks, Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: How does HA work?
On 24.10.2013 01:42, Chiradeep Vittal wrote: https://cwiki.apache.org/confluence/x/dwn8AQ Thanks! -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Enable SSL (https) for Cloudstack 4.2.0 management server
On 29.10.2013 09:02, Indra Pramana wrote: Dear all, Any documentation on how to enable SSL (https) for Cloudstack 4.2.0 management server? I am using Ubuntu 12.04.2. I tried to follow the instruction here, is it still valid? http://support.citrix.com/article/CTX132008 I have managed to come to the step of creating the PKS12 format keystore: Wow, that looks convoluted. Don't know if it would work, but my instinct is just "yum install mod_ssl" and proxy requests. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Docker?
Hi, Docker[1] seems to be all the rage these days and has breathed new life in LXC. Is there any interest whatsoever from the devs to support for it? I noticed it has already landed in the other stack. Lucian [1] - http://www.docker.io/ -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: CS 4.2 VM snapshot is not enabled for hypervisor type: KVM
On 31.10.2013 19:32, motty cruz wrote:
Hello,
I'm having issues with snapshot, when trying to take a snapshot I get
the
following error: "VM snapshot is not enabled for hypervisor type: VKM"
but I do have it enable in Global settings, am I missing a step?
kvm.snapshot.enabled = true,
Thanks,
Yes, I'm seeing the same. You can however snapshot the individual
root/data disks if you go in the Storage section which is a bit "wtf".
CC-ing "dev". Anyone knows why we can't snapshot KVM VMs from the
instance's "Details" menu? We should be able to perform live, disk-only
snapshots from this menu.
management-server.log says:
==> /var/log/cloudstack/management/management-server.log <==
2013-10-31 21:16:24,934 DEBUG [cloud.api.ApiServlet]
(catalina-exec-1:null) ===END=== 1.2.3.4 -- GET
command=listOsTypes&response=json&sessionkey=teaa9DZoOgNS38zwc%2FY2VwoUjoU%3D&_=1383254184625
2013-10-31 21:16:26,868 DEBUG [cloud.server.StatsCollector]
(StatsCollector-3:null) VmStatsCollector is running...
2013-10-31 21:16:26,947 DEBUG [agent.transport.Request]
(StatsCollector-3:null) Seq 1-1578762257: Received: { Ans: , MgmtId:
144350851819, via: 1, Ver: v1, Flags: 10, { GetVmStatsAnswer } }
2013-10-31 21:16:27,034 DEBUG [cloud.server.StatsCollector]
(StatsCollector-3:null) StorageCollector is running...
2013-10-31 21:16:27,103 DEBUG [agent.transport.Request]
(StatsCollector-3:null) Seq 3-852819979: Received: { Ans: , MgmtId:
144350851819, via: 3, Ver: v1, Flags: 10, { GetStorageStatsAnswer } }
2013-10-31 21:16:27,171 DEBUG [agent.transport.Request]
(StatsCollector-3:null) Seq 1-1578762258: Received: { Ans: , MgmtId:
144350851819, via: 1, Ver: v1, Flags: 10, { GetStorageStatsAnswer } }
2013-10-31 21:16:28,525 DEBUG [cloud.server.StatsCollector]
(StatsCollector-2:null) HostStatsCollector is running...
2013-10-31 21:16:28,551 DEBUG [agent.manager.AgentManagerImpl]
(AgentManager-Handler-13:null) SeqA 2-37914: Processing Seq 2-37914: {
Cmd , MgmtId: -1, via: 2, Ver: v1, Flags: 11,
[{"com.cloud.agent.api.ConsoleProxyLoadReportCommand":{"_proxyVmId":2,"_loadInfo":"{\n
\"connections\": []\n}","wait":0}}] }
2013-10-31 21:16:28,626 DEBUG [agent.manager.AgentManagerImpl]
(AgentManager-Handler-13:null) SeqA 2-37914: Sending Seq 2-37914: {
Ans: , MgmtId: 144350851819, via: 2, Ver: v1, Flags: 100010,
[{"com.cloud.agent.api.AgentControlAnswer":{"result":true,"wait":0}}] }
2013-10-31 21:16:29,147 DEBUG [agent.transport.Request]
(StatsCollector-2:null) Seq 1-1578762259: Received: { Ans: , MgmtId:
144350851819, via: 1, Ver: v1, Flags: 10, { GetHostStatsAnswer } }
2013-10-31 21:16:37,968 DEBUG [cloud.api.ApiServlet]
(catalina-exec-15:null) ===START=== 1.2.3.4 -- GET
command=createVMSnapshot&virtualmachineid=67dd260a-d951-427c-90f0-350f2f31114d&snapshotmemory=false&name=&response=json&sessionkey=teaa9DZoOgNS38zwc%2FY2VwoUjoU%3D&_=1383254197913
2013-10-31 21:16:37,991 INFO [cloud.api.ApiServer]
(catalina-exec-15:null) VM snapshot is not enabled for hypervisor type:
KVM
Lucian
--
Sent from the Delta quadrant using Borg technology!
Nux!
www.nux.ro
Re: [ACS 421] Upgrade testing
On 31.10.2013 02:51, Sudha Ponnaganti wrote: Hi All, Want to check if anyone is testing upgrades from prior versions to 4.2.1. As Release date is getting close, If there are gaps, would like to cover some scenarios in our lab. Thanks /sudha I'd love to test it, anyone got an RPM repo with it? Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Link between a snapshot and it's filename on disk
Hello, I'm don't seem to be able to match a snapshot from the UI with the file on disk in the secondary storage. I managed to find it by looking at timestamps, but otherwise there is no apparent (to me) link. Can anyone advise how I could find a snapshot's file on disk? For example I have a snapshot with name detached_fff_20131104155235 and ID bf08aea6-20d0-445f-b9e7-1039cea8d0a2. In the secondary storage the coresponding filename on disk is 622ddb9e-86f8-4040-883b-955f3cfdb936.vhd (found by timestamp match).. to me it wuld have made sense to have the filename match the ID. Pointers? Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Link between a snapshot and it's filename on disk
On 04.11.2013 16:30, Nux! wrote: Hello, I'm don't seem to be able to match a snapshot from the UI with the file on disk in the secondary storage. I managed to find it by looking at timestamps, but otherwise there is no apparent (to me) link. Can anyone advise how I could find a snapshot's file on disk? For example I have a snapshot with name detached_fff_20131104155235 and ID bf08aea6-20d0-445f-b9e7-1039cea8d0a2. In the secondary storage the coresponding filename on disk is 622ddb9e-86f8-4040-883b-955f3cfdb936.vhd (found by timestamp match).. to me it wuld have made sense to have the filename match the ID. Pointers? Lucian BTW, this is related to https://issues.apache.org/jira/browse/CLOUDSTACK-5035 I was advising Diogo to run "qemu-img info" on the file, if he could find it. :) Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: NFS vs Swift
On 05.11.2013 05:44, Damien Mannix wrote: HI All, We are currently running CloudStack 3.07 with NFS Secondary Storage which we have had continuing issues with. Has anyone migrated from NFS to swift? Pros / Cons – any tips would be most helpful. Thanks in advance The main problem I have with NFS is scalability, it will hold back the whole setup when it gets real big. Luckily there are workarounds without messing with swift or other s3 clones, one could use glusterfs' nfs interface + RR dns (we do this currently - though not for ACS - and it works well). I'd love to see native glusterfs support for secondary storage. My 2 strips of gold-pressed latinum -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: NFS vs Swift
On 05.11.2013 09:06, Damien Mannix wrote: Yes I have heard of glusterfs any tips, tricks - advice? Well, this is hardly the place, you should get on their ml, this is my first advice. Use CentOS/RHEL 6 if you can, use XFS as filesystem with 512b inodes. If you want speed you need 10G or Infiniband, otherwise you'll have to put up with sub 50 MB/s speeds on a mere 1Gbps network. HTH Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Bug? Should we allow detaching volumes when VMs have snapshots
On 07.11.2013 21:17, Mike Tutkowski wrote: Especially since it appears to have been working for XenServer and I'd hate to take that functionality away from people who might be using it. +100, removing functionality will not gain you any extra points; especially doing it for what is probably the least deployed hypervisor (quotation needed). -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Purpose of OS Type on templates
On 13.11.2013 10:11, Donal Lafferty wrote: Which hypervisors make use of OS type on a template's configuration? Hi Donal, KVM at least does, it exposes paravirt VirtIO devices if the OS type is known to support it. Xenserver also creates a PV or HVM vm based on the OS type. HTH Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Retreive VM snapshot size
On 13.11.2013 11:51, Antonio Petrocelli wrote: Hello, someone knows how to obtain the size of a VM snapshot with CLOUDSTACK 4.2 API? with this call _http://cloudstack.apache.org/docs/api/apidocs-4.2/domain_admin/listVMSnapshot.html_ size is not specified. This seems to be a known problem, someone else has already filed a bug: https://issues.apache.org/jira/browse/CLOUDSTACK-5035 Let's hope it gets some dev attention. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: CloudStack.next
On 12.11.2013 23:41, Steve Wilson wrote: Hi All, As we ramp towards freeze on 4.3 and start talking about 4.4, I thought it would be fun to queue up a discussion here on the list before Collab next week. What do you envision in the next MAJOR release of CloudStack? Call it 5.0 or whatever you like, but what would you like to see there? What would you change? What would you enhance? Are there big bets we should be placing as a community? Feel free to post any thoughts here and I'll look forward to talking to many of you in person at Collab next week. You are coming to Collab, right? My wishlist, based on not very intensive use so far: - better support for KVM (VM snapshots, SPICE support etc) - get rid of realhostip.com - dynamic size for the ROOT disk (so openstack templates can be used with minimal modifications) - good support for SDN (make VXLAN default etc) Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: CloudStack.next
On 18.11.2013 21:32, Chiradeep Vittal wrote: Realhostip.com is not required. See: http://support.citrix.com/article/CTX133468 But I agree that realhostip should not be made the default (default should be http). Can this be tweaked from Global Settings? Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [DISCUSS] Should we be releasing -beta releases?
On 14.05.2013 15:41, Chip Childers wrote: As a way to get more user feedback on our major feature releases, what does everyone think about releasing one or two -beta releases for each major feature release? Good idea! It would be great to get a feel of how things will look and work ahead of release! This could also allow people who want to do production deployments match their efforts with your development cycle (if said betas will be upgradable to release). Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
RE: There are private IP addresses allocated for this pod
On 22.06.2013 11:56, Nux! wrote: On 22.06.2013 11:36, Pranav Saxena wrote: Did you try deleting the physical network ? If not , try that . There is no network left (other than the one for the basic zone, which I need obviously). There's just the pod. Where is this private range in DB? I cold try to delete it from there and see if I can get rid of the pod. Tried to delete the pod from the DB directly, but I'm getting this: SQL query: DELETE FROM `cloud`.`host_pod_ref` WHERE `host_pod_ref`.`id` =2 MySQL said: Documentation #1451 - Cannot delete or update a parent row: a foreign key constraint fails (`cloud`.`cluster`, CONSTRAINT `fk_cluster__pod_id` FOREIGN KEY (`pod_id`) REFERENCES `host_pod_ref` (`id`)) CC-ing in dev@, maybe they can advise on how to remove a Pod from the DB directly. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
RE: There are private IP addresses allocated for this pod
On 22.06.2013 15:15, Nux! wrote: On 22.06.2013 11:56, Nux! wrote: On 22.06.2013 11:36, Pranav Saxena wrote: Did you try deleting the physical network ? If not , try that . There is no network left (other than the one for the basic zone, which I need obviously). There's just the pod. Where is this private range in DB? I cold try to delete it from there and see if I can get rid of the pod. Tried to delete the pod from the DB directly, but I'm getting this: SQL query: DELETE FROM `cloud`.`host_pod_ref` WHERE `host_pod_ref`.`id` =2 MySQL said: Documentation #1451 - Cannot delete or update a parent row: a foreign key constraint fails (`cloud`.`cluster`, CONSTRAINT `fk_cluster__pod_id` FOREIGN KEY (`pod_id`) REFERENCES `host_pod_ref` (`id`)) CC-ing in dev@, maybe they can advise on how to remove a Pod from the DB directly. Anyone? -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: There are private IP addresses allocated for this pod
On 24.06.2013 15:21, Daan Hoogland wrote: Nux, Have a look at the file create_schema.sql from the git repository. As far as I know there are no cascading deletes supported and as you started deleting you'll have to follow the relations yourself. a mysql workbench is going to be your best friend right now. 'host_pod_ref' is not ref'd from just the host and pod tables. Daan Thanks Daan, unfortunately had to rebuild the whole data base. I guess I'll have to leave the solution to this for next time it happens. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [DISCUSS] Should we be releasing -beta releases?
On 14.05.2013 15:41, Chip Childers wrote: As a way to get more user feedback on our major feature releases, what does everyone think about releasing one or two -beta releases for each major feature release? This might fall in line with some of the stated concerns about our release schedule (see [1]). I've stated a desire to be quicker about our releases (my vote was 4 months). I've also been saying quite publicly that we should never release if we know about upgrade issues (that's the cost of having actual users of our project, which I'm more than willing for us to pay). Perhaps -betaX releases would be helpful to get attention from the users to test the release (including upgrade paths). The stated assumption could be: -beta releases are not releases that can be upgraded *from*, but are intended to help support testing by end users that want to check the upcoming release against their expected feature set and upgrade path. I would see the first -beta-1 being released about 1 month after feature freeze. For example, for 4.2.0, it would be on 2013-06-30. I would only do a -beta-2 (or later) beta release if required due to testing results. I would also suggest that the -beta-* releases would *not* have any particular quality criteria (well... perhaps minimal, like blocking on issues that fundamentally make the software unstable). I'm not sure about my own proposal here, but I wanted to throw it out and see if any of you have feedback / thoughts. -chip [1] http://markmail.org/message/3ctdwor5hfbpa3vx +1 for beta releases, I was actually thinking of building some RPMS from source, want to get a flavour of 4.2 features, but not sure if I can be bothered with that. If I had some nightlies or betas available on cloudstack.apt-get.eu I'd definitely give it a go. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [DISCUSS] Should we be releasing -beta releases?
On 14.05.2013 15:41, Chip Childers wrote: As a way to get more user feedback on our major feature releases, what does everyone think about releasing one or two -beta releases for each major feature release? Hi, What has been decided? Will we see any 4.2 betas? Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [DISCUSS] Should we be releasing -beta releases?
On 31.07.2013 16:30, Chip Childers wrote: On Wed, Jul 31, 2013 at 12:04:50AM +0100, Nux! wrote: On 14.05.2013 15:41, Chip Childers wrote: As a way to get more user feedback on our major feature releases, what does everyone think about releasing one or two -beta releases for each major feature release? Hi, What has been decided? Will we see any 4.2 betas? Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro I think that we realized that the upgrade support problems are significant enough to make this difficult right now. Consider it an aspiration for the future. To be honest even "unsupported" betas might be good. I'd be willing to test betas even without upgradability to "stable", just to see what to expect, what's new, what's good, what's bad etc. I'm sure there are many like me. Sure, I can download and build it myself, but it would've been much more convenient to have some RPMs on cloudstack.apt-get.eu; plus, not everyone is comfortable building RPMs or from source etc. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [DISCUSS] Should we be releasing -beta releases?
On 05.08.2013 16:09, Daan Hoogland wrote: For this the jenkins builds might be suitable. I think if those are running on all active release branches. Some URLs could be real handy now. :) -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [DISCUSS] Should we be releasing -beta releases?
On 05.08.2013 16:48, Daan Hoogland wrote: the 4.1 build for rhel63 says: Apache Cloudstack RHEL 6.3 packages This project is currently disabled this one didn't happen to be the one you where looking for? Anyway look at http://jenkins.cloudstack.org and browse around. I thought you would find them there. Thanks, I'll have a look around there. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
What are the new features in 4.2?
Hello guys, I've been trying to find a definitive "new features" for 4.2, but failed. Can anyone point me to the right URL or make a summary of what's new? Thanks, Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: What are the new features in 4.2?
On 06.08.2013 18:52, Chiradeep Vittal wrote: I look at Developer -> Design Documents -> 4.2 3rd Party UI Plugin Ability to delete or archive Events and Alerts Woah, massive changes! I welcome them all. :D Thanks guys. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
IPv6 support
Hi, Should I expect to have working IPv6 (for the VMs) in 4.1 or 4.2? If not, in which version is this feature expected to land? Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
RE: IPv6 support
On 15.05.2013 18:21, Sangeetha Hariharan wrote: Hi, IPV6 feature is supported in 4.1 but it is experimental and UI is not available. Also you will need to use System Vm templates that have support for ipv6. Only Phase 1 from the FS - https://cwiki.apache.org/confluence/display/CLOUDSTACK/IPv6+support is supported. In 4.2 , there will be new System Vm templates that are made available that will have support for ipv6. Also UI will have support of IPV6 feature. -Thanks Sangeetha Thanks Sangeetha! Looking forward to the 4.2 Betas then. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: VXLAN problems
On 21.01.2014 05:25, Toshiaki Hatano wrote: Hi Nux, Marcus, It's FYI: There's compiled vxlan document in Jenkins. http://jenkins.buildacloud.org/job/build-docs-vxlan-master/ Bookmarked, thanks! -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Cloudstack 4.3 on Centos i686
On 21.01.2014 22:56, Prabhakaran Ganesan wrote: Hi Wido Thanks a lot for your response. I happened to have a i686 Centos host, hence the question. Let me upgrade the server to x86_64 and give it a try. At best you will be able to use a 32bit machine as a management server, (RH)EL 6 does not support KVM hypervisor on 32 bit. If you want to run a CentOS hypervisor you need x86_64. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Useless egress in SG zone?
Hi, I'm testing ACS 4.2 + XS 6.2 with Basic Zone and it kind of works great, but the Egress rules seem to be useless. Since all outgoing traffic is accepted by default, what is their purpose? Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Useless egress in SG zone?
On 25.01.2014 01:12, Marcus Sorensen wrote: Are you talking about the rules that ensure an instance can't bring up and use IP addresses that are not assigned to it? I'm not sure. Here's a pic: http://img.nux.ro/jC4b-Selection_015.png The anti-spoofing is working ok, supposedly, but I was expecting that either: 1 - egress is blocked by default, just like ingress, so just ports/addresses specified there can be accessed 2 - less orthodox, but since we allow all outgoing by default for a VM then make this is a blacklist instead of a whitelist, ie ports/addresses specified here cannot be accessed Do I make any sense? Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Feature request: scale up for KVM
Hi, I'd like to see this feature available for KVM, ie live changing offering. Any chance someone can make it come true? Opened https://issues.apache.org/jira/browse/CLOUDSTACK-5945 for it. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Feature request: scale up for KVM
On 25.01.2014 04:27, Marcus Sorensen wrote: There are some issues with it. Last I checked, you can hotplug CPUs, but you can't take them away (at least not for Linux guest os). Also, you can scale memory, but it relies on balloon driver, which can simply be unloaded by the guest to have access to all memory as the driver inside the guest handles the memory size. I believe there was work on real memory hotplug, but it might be awhile before that makes it to standard guest kernels. Ah, okay, I didnt know the balloon driver can be fooled like this. I guess we need to wait for stuff to get fixed first, thanks! Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: XenServer VM live scales the CPU, but not the RAM
Hi, Copying dev@, maybe someone has an idea for the below? On 27.01.2014 10:28, Nux! wrote: Hi, I'm testing Xenserver live scale-up feature, and it almost works. I'm using a Centos 6 64bit PV guest and trying to upgrade from an offering of 2 CPU and 512MB RAM to 3 CPU and 1024MB RAM. The CPUs scale up nicely, but not the memory, it only scales from 483 MB (as reported by `free -m`) to 532 MB and in XenCenter the memory is also not what it should: http://img.nux.ro/mK7-vmmem116.png What am I doing wrong? Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
RE: XenServer VM live scales the CPU, but not the RAM
On 27.01.2014 15:21, Prashant Kumar Mishra wrote: Can you share xen server version and logs . Hi, Xenserver is 6.2, no hotfixes applied. Which log files are you interested in? (sorry, not terribly familiar with Xenserver) Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: KVM memory overprovision breaking me
On 28.01.2014 00:49, Marcus Sorensen wrote: So... I tried to use memory overcommit on KVM this week, and it blew up in my face. Apparently it's configured such that if I have a Service Offering of 4G, and I set memory overprovisioning to 2:1, the guest only actually gets configured with 2G. That's not how overprovisioning is supposed to work, IMO. Here's a vm definition with a 3:1 mem overprovision setting, which ensures that system vms don't work: 262144 87040 Note currentMemory needs to be manually tuned if I ever want the vm to use/see more. This is more for live scaling (which is also broken because the guest could just rmmod virtio-balloon and see everything). I'd like to just rip out the code that is setting ballooning feature based on overprovisioning factor, but perhaps there was a reason this was done. From my point of view, if I give someone a service offering that says 4G, it should provide 4G, and if I can do memory deduplication on the backend to overprovision that's up to me to do. Overprovisioning should not be a divider on all service offerings. Wow! I also thought, heck, KSM & thin qcows for the win! If overprovisioning really "works" as you described then it can't possibly be used for any commercial offering ... This needs to get fixed.. Too late to see this in 4.3? -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: KVM memory overprovision breaking me
On 28.01.2014 06:57, Marcus Sorensen wrote: Actually, does anyone have objections to adding an agent.properties tunable to disable balloon? That would allow anyone who wants to overcommit to do so, without messing with the existing FS, and the calculations for allocation at least are all accurate. And I won't have to maintain another local patch. I just can't give someone a 4G VM and have it only show 2-1G inside, just like I wouldn't give someone a 1T thin-provisioned disk and only show them 100G. I know KSM will give me X memory dedup for my workload, and I'd prefer to straight up over allocate VMs to the guests based on that. Balloon just doesn't work anyway for most, only if the customers are internal and agree not to meddle with it. I think that goes for all hypervisors. +100 on the above! You can count of me for testing. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: KVM memory overprovision breaking me
On 28.01.2014 09:56, Bharat Kumar wrote: Hi, The calculations when overcommitting are on a per VM basis. I am not sure if these will be valid when using KSM. IMO KSM is applicable to a set of VMs not a single VM. so how do we keep track of how much free memory is actually available on the host. This was not problem in case of ballooning as we knew upfront about the limits that we want to set on a per VM basis. The problem with ballooning is that it depends in you having exclusive root access on the VMs and on running an agent inside the VMs (it doesnt even work for all OSes afaik). This is a very particular situation and for people who want to sell cloudstack to the public it's a no-go. I would be very happy if ACS allowed me to set an overprovisioning factor and then let me use KSM without having to mess with the VMs. Available memory on the hypervisor could be tracked by standard linux utils via the agent (`free`?). -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Useless egress in SG zone?
On 28.01.2014 05:20, Jayapal Reddy Uradi wrote: Hi Nux, 1. By default we are allowing egress in SG. 2. But when you configure any rule in egress, it allows ONLY configured rule traffic and other traffic will be BLOCKED. If admin wants allow to only specific ports/addresses this can be done by configuring SG egress rules. In my firewalls, the default egress is allow for trusted networks. Thanks, Jayapal Thanks a lot Jayapal, this makes sense. What was confusing me is that in a Basic zone (hence SG) with Xenserver this is not working. I can confirm it is working in an adv zone + SG with KVM. Regards, Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: KVM memory overprovision breaking me
On 28.01.2014 17:49, Marcus Sorensen wrote: Commit 1530c162e58abd8e572a89d2a2706de00bd49e26 in branch refs/heads/master from [~mlsorensen] [ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=1530c16 ] CLOUDSTACK-5968 create vm.memballoon.disable agent parameter How do I lobby for it? :) Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Cherry-pick request for 4.3 please, re memory overprovisioning
Hello, Kindly add this to 4.3 https://issues.apache.org/jira/browse/CLOUDSTACK-5968 It enables memory overprovisioning without modifying VM internal settings, quite critical for cloud sellers. Thanks! -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Add new KVM hypervisor host into CloudStack
On 30.01.2014 08:57, Indra Pramana wrote: Hi, It's time for me to add new KVM hypervisor host into our CloudStack and I read these requirements: http://cloudstack.apache.org/docs/en-US/Apache_CloudStack/4.2.0/html/Installation_Guide/hypervisor-kvm-install-flow.html - Same distribution version, is it referring to same OS and version? E.g. Ubuntu 12.04 (Precise)? - All hosts must be homogenous, does it mean that the new server has to be exactly the same chassis, motherboard, type of CPU and RAM? Can I use a different type of chassis and motherboard, but use the same type of CPU? Thank you. Yes, I think the CPU is the important bit here and imho it should be equal or better than what you have if you want certain stuff like live migration to work. This is especially important if you use host-passthrough for CPU type. I do not think anyone expect you to have indentical servers throughout your setup; it's normal to bring in better servers as time passes. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
RE: [VOTE] Apache CloudStack 4.3.0 (second round)
On 29.01.2014 22:52, Geoff Higginbottom wrote: -1 I am unable to manage Firewall, Port Forwarding or Load Balancing rules via the UI, the interfaces no longer exist. I did create a Port Forwarding Rule via the API successfully to prove it's just a UI issue. VMs assigned to VPC's fail to deploy and are then deleted. I was unable to complete my remaining tests as these required a working VPC and the ability to create Firewall, Port Forwarding and Load Balancing Rules for Isolated Networks. Yeah, there are some problems with the new UI. I've also reported some issues: https://issues.apache.org/jira/browse/CLOUDSTACK-5906 https://issues.apache.org/jira/browse/CLOUDSTACK-5907 (not UI related) -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [RFC] adding volume provisioning method option
On 31.01.2014 20:24, Yoshikazu Nojima wrote: Afternoon All, Is there anyone working on adding volume provisioning method option? As you know, thin provisioning of a volume save consumption of a storage, and fat provisioning improves IOPS performance. Especially, Qcow2 can save storage consumption and achive relatively better performance than default by provisioning a volume with an option "preallocation=metadata", which makes an image file a sparse file. http://itscblog.tamu.edu/improve-disk-io-performance-in-kvm/#.232---preallocation Any thoughts about this? If it is ok, I will write a feature specification on confluence and start implementation. Regards, Noji Yoshikazu Nojima Hello, I thought preallocation=metadata is common practice since years. Now I find out ACS doesn't actually use it? If so, this is really _bad_ and needs to be fixed ASAP... Thanks Noji -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [RFC] adding volume provisioning method option
On 01.02.2014 07:14, Marcus wrote: Oh yes, and storage overprovisioning doesn't currently work for KVM storage types, but it's a relatively simple fix: Let me know when you need someone to ask for a cherry pick. :-) Qcow files scream overprovisioning. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [RFC] adding volume provisioning method option
On 01.02.2014 06:43, Marcus wrote: [root@server ~]# qemu-img create -f qcow2 -o preallocation=metadata imagepre.qcow2 10G Formatting 'imagepre.qcow2', fmt=qcow2 size=10737418240 encryption=off cluster_size=65536 preallocation='metadata' [root@server ~]# qemu info imagepre.qcow2 -bash: qemu: command not found [root@server ~]# qemu-img info imagepre.qcow2 image: imagepre.qcow2 file format: qcow2 virtual size: 10G (10737418240 bytes) disk size: 1.7M cluster_size: 65536 preallocation=metadata leaves disk as 1.7M, it's also sparse, but a bit bigger. I think just a global switch might be enough given the above. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: CLVM volumes shown under the wrong hypervisor type (OVM)
On 19.01.2014 18:00, Nux! wrote: Hi, Opened https://issues.apache.org/jira/browse/CLOUDSTACK-5907 for this. There doesn't seem to be any lack of functionality with these volumes, but it seems wrong to display OVM there instead of KVM and may cause other issues in the future. Does anyone have any idea why this is happening? -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
RE: [VOTE] Apache CloudStack 4.3.0 (third round)
On 06.02.2014 01:52, Animesh Chaturvedi wrote: Folks I need more feedback on this RC so that we can hopefully have the next RC as the last one. If you have not played / tested with this RC please do so now and share your feedback. I have some time for testing in the upcoming weekend, are these packages valid for this purpose? http://jenkins.buildacloud.org/view/4.3/job/cloudstack-4.3-package-rpm/lastStableBuild/ (#381) -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
RE: [VOTE] Apache CloudStack 4.3.0 (third round)
On 06.02.2014 02:21, Animesh Chaturvedi wrote: You should use the RC3 artifact in VOTE email https://dist.apache.org/repos/dist/dev/cloudstack/4.3.0/ Ack. RPMs here should someone else want to test without building them http://tmp.nux.ro/cloudsnap430_4308/ Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Location of the 4.3 System VM Templates
On 23.01.2014 14:39, Wido den Hollander wrote: On 01/23/2014 03:35 PM, David Nalley wrote: I'll respectfully disagree. We need to find a place for systemVMs whether that means classifying them as convenience binaries or putting them on Wido's box, or S3. I manually sync the images now. Sometimes I download some images to this location: http://cloudstack.apt-get.eu/systemvm/ It'd be nice if you could put the 4.3 ones there, too. People already know the location, they can browse it (unlike s3, hate that), for me it's a one stop source; plus it also gets mirrored[1]. [1] - http://mirrors.coreix.net/cloudstack.apt-get.eu/ -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
DB upgrade 4.2.1 - 4.3 fails
Just tested this today, it happens consistently in my setup. Opened https://issues.apache.org/jira/browse/CLOUDSTACK-6061 with plenty of details. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
RE: DB upgrade 4.2.1 - 4.3 fails
On 07.02.2014 19:01, Animesh Chaturvedi wrote: Nux you had a similar issue for previous RC https://issues.apache.org/jira/browse/CLOUDSTACK-5902 which was resolved as not a problem I know, but this time it's persistent. I think it may be related to the installation of the new system template. As specified in the bug, at what time must the new template be installed? I did it while still in running 4.2.1 and no luck, did it after, still no luck. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
RE: DB upgrade 4.2.1 - 4.3 fails
On 07.02.2014 19:47, Animesh Chaturvedi wrote: Did you review Alena's response? Of course, check the issue https://issues.apache.org/jira/browse/CLOUDSTACK-6061 -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: DB upgrade 4.2.1 - 4.3 fails
On 07.02.2014 20:55, Alena Prokharchyk wrote: See my reply to the bug. The new system vm template should be registered through CS API registerTemplate (and named like Install Guide requests), not by running template install script. I have corrected that, but the upgrade still fails. Issue updated with logs. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: DB upgrade 4.2.1 - 4.3 fails
On 07.02.2014 21:50, Alena Prokharchyk wrote: Nux, During the rollback, you just applied 4.2.1 db dump. That doesn¹t drop the tables that were inserted by partial 4.2.14.3 upgrade. Refer to my last comment to the bug for the rollback steps. Yes, that was it! On to more testing now, thanks! -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [VOTE] Apache CloudStack 4.3.0 (fourth round)
On 08.02.2014 07:12, Marcus wrote: +1 (binding) Built RPMs on CentOS 6.5, deployed advanced zone, vpc, network, installed template, launched VM on local storage, launched VM on NFS storage, launched VM on CLVM storage. Marcus, aren't you seeing the OVM issue with CLVM, too? https://issues.apache.org/jira/browse/CLOUDSTACK-5907 -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [VOTE] Apache CloudStack 4.3.0 (fourth round)
On 08.02.2014 10:00, Nux! wrote: On 08.02.2014 07:12, Marcus wrote: +1 (binding) Built RPMs on CentOS 6.5, deployed advanced zone, vpc, network, installed template, launched VM on local storage, launched VM on NFS storage, launched VM on CLVM storage. Marcus, aren't you seeing the OVM issue with CLVM, too? https://issues.apache.org/jira/browse/CLOUDSTACK-5907 Please ignore me, no longer happens with rev 4335. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [VOTE] Apache CloudStack 4.3.0 (fourth round)
On 08.02.2014 00:51, Animesh Chaturvedi wrote: Hi All, I've created a 4.3.0 release, with the following artifacts up for a vote: +1 from me, upgrade from 4.2.1 works fine. Tested with CentOS, KVM, NFS, CLVM, Adv+SG. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: What does this UI thing do? "Root disk size"
On 20.01.2014 23:11, Mike Tutkowski wrote: So, is this GUI field going to be in 4.3, but not functional? If so, that seems a little strange. Perhaps it should be removed. BTW, guys, this thing still shows up in the UI. Should be removed, creates confusion and false hope. :-) -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Failed to get rule logs, better luck next time
Hi, At some point during my migration from 4.2.1 to 4.3 I started getting these on the Cloudstack KVM HV: Failed to get rule logs, better luck next time! It happens when this is run: /usr/share/cloudstack-common/scripts/vm/network/security_group.py get_rule_logs_for_vms Can someone help me investigate this further? I think it may be trying to get the rules for some long gone VM, but not sure which. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Weird HA issue in 4.3
Hi, I'm testing HA in 4.3 and for the most part it works ok (if I kill a HV the VMs are started on the others), however I have an issue in that when killing a VM (powering it off from console) Cloudstack doesn't turn it back on. It marks it as "stopped" in the UI after a while, but doesn't start it back on. Is this the expected behaviour? I no longer have 4.2.1 installed, but from what I remember it was doing the right thing and starting the VM. Thoughts? -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Weird HA issue in 4.3
On 09.02.2014 17:04, Marcus wrote: Oh, sorry, I see you mean you turn it off outside of cloudstack. If CS checks the server and notices that a vm is no longer running, and it thought the vm was running, it should probably start it. Yep, I stopped it by issuing a poweroff via console. I'll try to test this some more because it doesn't look like it should happen this way. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Weird HA issue in 4.3
On 09.02.2014 17:21, Nux! wrote: On 09.02.2014 17:04, Marcus wrote: Oh, sorry, I see you mean you turn it off outside of cloudstack. If CS checks the server and notices that a vm is no longer running, and it thought the vm was running, it should probably start it. Yep, I stopped it by issuing a poweroff via console. I'll try to test this some more because it doesn't look like it should happen this way. Ok, so this happens every time so far. I stop the VM from within via console or ssh, the management server spits out this http://fpaste.org/75667/67633139/raw/ and the UI keeps showing the VM as running. I've stopped the VM ~30 minutes ago and no change. I'll open an issue. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Weird HA issue in 4.3
On 09.02.2014 18:29, Nux! wrote: On 09.02.2014 17:21, Nux! wrote: On 09.02.2014 17:04, Marcus wrote: Oh, sorry, I see you mean you turn it off outside of cloudstack. If CS checks the server and notices that a vm is no longer running, and it thought the vm was running, it should probably start it. Yep, I stopped it by issuing a poweroff via console. I'll try to test this some more because it doesn't look like it should happen this way. Ok, so this happens every time so far. I stop the VM from within via console or ssh, the management server spits out this http://fpaste.org/75667/67633139/raw/ and the UI keeps showing the VM as running. I've stopped the VM ~30 minutes ago and no change. I'll open an issue. Created https://issues.apache.org/jira/browse/CLOUDSTACK-6065 -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [VOTE] Apache CloudStack 4.3.0 (fourth round)
On 08.02.2014 10:56, Nux! wrote: On 08.02.2014 00:51, Animesh Chaturvedi wrote: Hi All, I've created a 4.3.0 release, with the following artifacts up for a vote: +1 from me, upgrade from 4.2.1 works fine. Tested with CentOS, KVM, NFS, CLVM, Adv+SG. There might be a problem with the HA engine, please see https://issues.apache.org/jira/browse/CLOUDSTACK-6065 -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Missing cidrlist in 4.3 adv zone firewall
Hi, It's the first time I'm testing firewall in 4.3 Advanced zone (without SG) so please let me know if I'm missing something obvious; I notice the cidrlist is missing from the rules, both in UI and in cloudmonkey. If I create the rule from cloudmoneky it also doesn't register a cidrlist, so it doesn't seem to be UI's fault. This is what I see in the logs http://fpaste.org/75819/39203643/ when I create a rule. Anyone else experiencing this? See http://img.nux.ro/3Kk-Selection_050.png mycloudmonkey > list firewallrules id=835dfc08-beab-458a-9c30-6b0b2b11f201 count = 1 firewallrule: id = 835dfc08-beab-458a-9c30-6b0b2b11f201 cidrlist = endport = 65535 ipaddress = 172.16.72.212 ipaddressid = f481629a-deb6-4413-b253-e8e98d8a303a networkid = c615df7c-3ea3-4138-a83c-d848e20fe1f6 protocol = tcp startport = 1 state = Active tags: -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Missing cidrlist in 4.3 adv zone firewall
On 10.02.2014 13:28, Jayapal Reddy Uradi wrote: Hi, Cidr showing correctly in our setup. Is cidr stored in db table firewall_rules_cidrs ? Jayapal, I can see the CIDRs in that table. I also checked inside the VR and while iptables does have rules for the ports I mentioned, the CIDRs are missing, too. Do note this is an upgrade from 4.2.1. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
RE: [VOTE] Apache CloudStack 4.3.0 (fourth round)
On 10.02.2014 21:17, Edison Su wrote: +1(binding). Following the test procedure, and test it on devcloud, it works as it is. Is it okay to leave the HA issue for subsequent RCs? https://issues.apache.org/jira/browse/CLOUDSTACK-6065 -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
RE: [VOTE] Apache CloudStack 4.3.0 (fourth round)
On 11.02.2014 00:50, Animesh Chaturvedi wrote: -Original Message- From: Nux! [mailto:n...@li.nux.ro] Sent: Monday, February 10, 2014 3:41 PM To: dev@cloudstack.apache.org Subject: RE: [VOTE] Apache CloudStack 4.3.0 (fourth round) Is it okay to leave the HA issue for subsequent RCs? https://issues.apache.org/jira/browse/CLOUDSTACK-6065 [Animesh] I will let folks weigh in on whether this should block the release. Does it affects other hypervisors besides KVM? Seems specific to KVM agent from what I understand in Sadhu's comments. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Network Interface Files with bootproto DHCP
On 11.02.2014 12:37, Gaurav Aradhye wrote: Hello all, I am trying to add network interface file for a secondary IP through code. The simplest way would be copying the file for the primary IP and changing required values accordingly. But it seems that as the boot protocol value is DHCP, the values such as ipaddress, netmask, gateway are not stored in the ifcfg-eth0 file. Are these values not stored anywhere on the VM? Also, what would be other way to add network interface file for secondary IP? I believe CloudStack doesn't do this itself after acquiring secondary IP. Regards, Gaurav Gaurav, It's not ACS's job to the set the 2nd IP on the VM. You can easily do it from within. This is sysadmin job. The "easiest", but not standard way of doing it is just adding something like this in rc.local: ip a a 1.2.3.4/32 dev eth0 The better way to do it is just create another ifcfg file (assuming RHEL/CentOS) that looks like this: DEVICE=eth0:1 IPADDR=1.2.3.4 PREFIX=32 ONBOOT=yes HTH Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [VOTE] Apache CloudStack 4.3.0 (fourth round)
On 08.02.2014 10:56, Nux! wrote: On 08.02.2014 00:51, Animesh Chaturvedi wrote: Hi All, I've created a 4.3.0 release, with the following artifacts up for a vote: +1 from me, upgrade from 4.2.1 works fine. Tested with CentOS, KVM, NFS, CLVM, Adv+SG. I'm also making it -1, based on https://issues.apache.org/jira/browse/CLOUDSTACK-6065 - broken HA on KVM https://issues.apache.org/jira/browse/CLOUDSTACK-6083 - missing cidrs in firewall rules after upgrade from 4.2.1 -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Missing cidrlist in 4.3 adv zone firewall
On 12.02.2014 13:12, Jayapal Reddy Uradi wrote: Hi Nux, I found this issue in upgraded setup with external firewall srx. I looking into this issue. Thanks, Jayapal No, thank you :) -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [VOTE] Apache CloudStack 4.3.0 (fourth round) UPDATE
On 13.02.2014 05:28, Animesh Chaturvedi wrote: 1. CLOUDSTACK-6083 : Missing CIDRs in firewall rules after upgrade => Waiting on Jayapal to confirm and update the list with findings Jayapal's fix works for me. 3. CLOUDSTACK-6065: KVM HA not working for shutdown VMs => While an issue but my understanding is it is a corner scenario only affecting KVM and can be released as a known issue Well, it's effectively breaking HA as defined in the docs for a main hv type and as it is I can't use 4.3 in production. Is there a fix in sight? Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
RE: [VOTE] Apache CloudStack 4.3.0 (fourth round) UPDATE
On 13.02.2014 23:01, Animesh Chaturvedi wrote: 3. CLOUDSTACK-6065: KVM HA not working for shutdown VMs => While an issue but my understanding is it is a corner scenario only affecting KVM and can be released as a known issue Well, it's effectively breaking HA as defined in the docs for a main hv type and as it is I can't use 4.3 in production. Is there a fix in sight? [Animesh] Following up with Kelven if he can check on it [Animesh] Nux Kelven fixed this particular issues in 4.3-forward branch. Can you test it out? Wunderbar! I'll test it tomorrow. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [VOTE] Apache CloudStack 4.3.0 (fifth round)
On 14.02.2014 06:24, Animesh Chaturvedi wrote: Hi All, I've created a 4.3.0 release, with the following artifacts up for a vote: +1 from me, the major issues I reported previously (broken HA & fw) have been fixed. Good job! Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
How to modify the (global) default security group?
Hello, In the light of recent UDP based attacks (NTP/DNS/SNMP reflection/amplification) I was thinking to hack the global default security group so that at least NTP and SNMP are protected in the whole cloud. By default the security groups come with everything blocked on ingress and all accepted on egress, I'd like to tune this a bit, have some ports open by default and some blocked. Any suggestions? Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: UI Latency
On 13.02.2014 18:04, John Kinsella wrote: Just posted http://theresnomon.co/2014/improving-cloudstack-ui-performance/ - open to any comments. That's great, some people are already using Apache HTTPD for SSL termination, this is a great bonus. Thanks for sharing. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Primary storage maintenance doesn't trigger VM shutdown in 4.3
Hi, It looks like putting the primary storage in maintenance doesn't trigger the VM shutdown as described in the docs Opened https://issues.apache.org/jira/browse/CLOUDSTACK-6121 with some more details. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Not_Able_To_Add_Disk_To_Windows_VM
On 15.02.2014 10:42, Mardan Raghuwanshi wrote: Hello All, I am running a windows VM on XEN hypervisor through cloudstack management server. I am trying to add a disk to windows VM but its throwing a error message as... *Failed to attach volume: superfastvolume to VM: win2k8ft1; You attempted an operation that requires PV drivers to be installed on the VM. Please install them by inserting xen-pv-drv.iso*. I am trying to find these drivers but i couldn't find out, Please help me where i can find these PV drivers. If you click on "attach iso" to VM, you should be offered to mount the xs-tools.iso. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [VOTE] Apache CloudStack 4.3.0 (sixth round)
On 19.02.2014 10:55, Hugo Trippaers wrote: -1 (binding) I’m trying to upgrade an existing cloudstack installation from version 4.2.1 to version 4.3 and this is flinging because the database checker can’t find the correct systemvm templates for the 4.3 release. I found the release notes at http://apache-cloudstack-release-notes.readthedocs.org/en/latest/rnotes.html with the upgrade instructions, but these point to wrong download locations and a wrong version number for the systemvm template (systemvm-xenserver-4.2). So i think that currently users can’t upgrade from a previous release to the new release. I successfully upgraded from 4.2.1 (CentOS/KVM) to 4.3, but before that I had to register the sysvm template from http://jenkins.buildacloud.org/view/4.3/job/cloudstack-4.3-systemvm/ and all went smoothly. HTH -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
RE: [VOTE] Apache CloudStack 4.3.0 (sixth round)
On 20.02.2014 00:10, Animesh Chaturvedi wrote: Folks I want to clear up the confusion on system templates for 4.3 I have noticed the system vms come up with 128 MB ram, isn't this a bit too little for 64bit machines? Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
RE: [VOTE] Apache CloudStack 4.3.0 (sixth round)
On 20.02.2014 11:58, Sateesh Chodapuneedi wrote: -Original Message- From: Nux! [mailto:n...@li.nux.ro] Sent: 20 February 2014 17:05 To: dev@cloudstack.apache.org Subject: RE: [VOTE] Apache CloudStack 4.3.0 (sixth round) On 20.02.2014 00:10, Animesh Chaturvedi wrote: Folks I want to clear up the confusion on system templates for 4.3 I have noticed the system vms come up with 128 MB ram, isn't this a bit too little for 64bit machines? The memory requirements of each system vm varies based on the load of services running. From my setup I can see only elasticloadbalancer has 128MB. Can you please let me know for which system vm did you observe 128MB RAM? The virtual router, but I think I am wrong now that I checked again, it is indeed 128 MB, but somehow I ended up with 32bit version of the template. Will need to redeploy systemvms with the 64bit template.. Sorry for the noise. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [DISCUSS] Browser-based template and volume uploads
On 19.02.2014 22:12, Chiradeep Vittal wrote: Hi all, The fact that one has to have a HTTP server to host templates before the templates can be imported into secondary storage is an inconvenience. I am not sure how much of an inconvenience it is, but I was asked to detail the technical architecture, if it needed to be supported. Certainly, I can imagine for folks coming in contact with ACS for the first time, it might seem onerous to set up another web server. I've sketched a strawman proposal [1] to support browser-based uploads of templates and volumes to secondary storage. It is not yet a PROPOSAL since it is still raw (and I do not have time to work on this immediately). Please comment on this. [1] https://cwiki.apache.org/confluence/x/jY5cAg A small request, can we enhance this feature so that it will generate multiple templates with different storage? For example now when I add a Linux template it has to be in different sizes: 10GB, 20GB, 40GB, 100GB, 200GB etc. Another round of templates for say Windows, another for eg Linux+CPanel and so on and so forth. I end up adding tons of templates (really one of my biggest pet peeves with ACS) whereas with the same setup in openstack or onapp I just need to add 3. -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [DISCUSS] Browser-based template and volume uploads
On 20.02.2014 23:04, Chiradeep Vittal wrote: Nux, that's an entirely different problem. Care to raise a feature request? It's been here for a while: https://issues.apache.org/jira/browse/CLOUDSTACK-5305 It's a major PITA for me (and I imagine for anyone else who wants to sell cloud publicly). -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [DISCUSS] Browser-based template and volume uploads
On 21.02.2014 06:01, Marcus wrote: If someone gets around to it they can add in root resize support via the patch I've posted to the list once or twice. I can maybe look at updating it for master and testing if I can find some time. It's a very short patch, but the testing is what I don't have time for at the moment, more of a priority thing. That would eliminate you having to upload multiple pre-sized templates. I would love to see this in the next version. What needs to be done? -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
status of 4.3?
Hi, Is 4.3 stuck because of the mysql licensing? Need to deploy ACS and I would've gone for the last RC which passed all my tests, but not unless there are going to be any dramatic last minute changes. Thoughts? Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
[4.4] Feature request: root resize (FAO Marcus)
Hi, Having talked about this issue recently, I'd like to request this feature so as to avoid template duplication. Marcus, you said you have a working patch, if you get it to apply to 4.4 I will invest time in testing it. I have opened an issue with rationale, example etc https://issues.apache.org/jira/browse/CLOUDSTACK-6181 Don't know what else I should be doing for a feature request. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: vhd-util on 64 bit system
On 27.02.2014 13:33, Daan Hoogland wrote: H, I am running the svm build scripts on a 64 bit centos64. I ggot stuck when the downloaded vhd-util asked for libuuid which i have installed but turns out to be 64 bit. Is there source or a 64 bit vhd-util lying around. I found a github repo [1] which contains source. Is this the vhd-util we need with cloudstack? [1] https://github.com/xapi-project/blktap/tree/master/vhd You should be able to install the 32bit version of that. What are the exact errors you are getting? Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: vhd-util on 64 bit system
On 27.02.2014 13:44, Daan Hoogland wrote: I have plunged into a dependency circus. at the moment it is libuuid. It was ld-linux. I dread what is coming. Don't dread it, just use yum provides */libuuid.blah I just built this on EL 6.5 + EPEL: http://tmp.nux.ro/3KR-vhd This is what I have installed: rpm -qa|grep libuuid libuuid-2.17.2-12.14.el6.x86_64 libuuid-2.17.2-12.14.el6.i686 libuuid-devel-2.17.2-12.14.el6.x86_64 -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: vhd-util on 64 bit system
On 27.02.2014 13:54, Daan Hoogland wrote:
I loaded the 32 bit libs and checked with ldd. All right so far. It
doesn't cantain the command convert however, which is needed to create
the xenserver templates.
I'll try your version Lucian,
It doesn't have the convert either,
./vhd-util
usage: vhd-util COMMAND [OPTIONS]
COMMAND := { create | snapshot | query | read | set | repair | resize |
fill | coalesce | modify | scan | check | revert }
I use qemu-img to convert qcow or raw to vhd (vpc)..
Lucian
--
Sent from the Delta quadrant using Borg technology!
Nux!
www.nux.ro
Re: [RFC] KVM snapshot support on RHEL(6.5 <= vesion < 7)
On 27.02.2014 23:15, Edison Su wrote: As we discussed[1][2] before, in ACS release, by default, there is no live volume snapshot for KVM, due to starting from 6.5, qemu-img can't recognize "-s", which used by CloudStack to backup snapshot. One possible solution, to bring the function back, is to build a separate qemu-img rpm from 6.5 with "-s" support, we can name the rpm as "cloudstack-qemu-img-extra" or whatever, and make sure this rpm won't overwrite anything installed on the system, so that won't break RHEL supportability(if anybody care). The changes we need: 1. Build a RPM with "-s" support 2. Push the RPM to a public repo 3. During KVM agent installation procedure, need to install above RPM after KVM agent installed. This step can be either manually, or automatically. Any comments on above steps? [1] http://markmail.org/message/b57vbbvjzp5g2562 [2] http://www.nux.ro/archive/2014/01/Taking_KVM_volume_snapshots_with_Cloudstack_4_2_on_CentOS_6_5.html To be honest I'm quite comfortable with leveraging manually the "cloud-qemu-img" functionality in ACS as per my blog post, of course large deployments can script this via ansible/puppet/etc; shipping it via RPM might require a bit of effort but it's doable, too. This will not be an issue with EL7 AFAIK. I don't know what got into RH with this change, it also was not fair play on the oVirt people. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Changing the SSL certificate for my own realhostip
Hi, I'm trying to implement my own realhostip and I have a problem with adding the certificate. What I have is a Comodo wildcard ca_bundle, crt and key in pem format (for use with Apache HTTPD) and Cloudstack is asking for "X.509 compliant SSL certificate" and "PKCS#8 Private Key". I have never used these formats and seem to be popular within the java world. I tried converting what I had from pem to der (pkcs8?) format[1], but what I get is some binary .der file that's "illegible" and can't really be copy-pased into the UI. Can someone advise on what steps I should take to get my SSL certs into ACS? I'm on 4.3.0 rev 4440. Regards, Lucian [1] cat ca_bundle.crt certificate.crt > certificate.pem openssl x509 -outform der -in certificate.pem -out certificate.der -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Changing the SSL certificate for my own realhostip
On 28.02.2014 12:59, Wei ZHOU wrote: There are three text fields, (1) copy certificate.crt and ca_bundle.crt, paste into the first text fields. (2) copy your key to the second (3) input your domain suffix This I have already done and while I got no error in the UI, when trying to open up the console for a VM I would get a nasty SSL error which I forgot to note down and have since reinstalled, but I'll try it again shortly. Thanks! Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Changing the SSL certificate for my own realhostip
On 28.02.2014 13:45, Wei ZHOU wrote: Did you use firefox? Yes, does it matter? -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Changing the SSL certificate for my own realhostip
On 28.02.2014 13:57, Wei ZHOU wrote: Yes. You need to download the Intermediate CA and import it into Firefox. URL: https://support.comodo.com/index.php?_m=downloads&_a=viewdownload&downloaditemid=122&nav=0,1 Firefox: Tools > Options > Advanced > Certificates/Encryption: View Certificates > Authorities > Import Firefox already includes this apparently, it would have been a problem if it hadn't (might as well go with self-signed). Anyway, after following your initial instructions this is What I'm getting when trying to open the console for a VM: "Secure Connection Failed An error occurred during a connection to 1-2-3-4.mycloud.tld. A PKCS #11 module returned CKR_DEVICE_ERROR, indicating that a problem has occurred with the token or slot. (Error code: sec_error_pkcs11_device_error) The page you are trying to view cannot be shown because the authenticity of the received data could not be verified. Please contact the web site owners to inform them of this problem. Alternatively, use the command found in the help menu to report this broken site." I also restarted the management server as well as the recreating the console vm, to no avail. Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Changing the SSL certificate for my own realhostip
On 28.02.2014 14:32, Wei ZHOU wrote: Can you visit the console by IE or Chrome? Hm, indeed, I tried Chromium and Opera and they do not complain. Only Firefox and Seamonkey: The certificate is not trusted because no issuer chain was provided. At least we know the certificates get imported in ACS just fine and this is not an issue. I must have forgotten to import some weird intermediate crt. Thanks! -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [DISCUSS] realhostip.com going away
On 28.02.2014 20:27, John Kinsella wrote: Folks: Recently the PMC was informed that the realhostip.com DNS service that ACS currently uses by default as part of the console proxy will be disbanded this summer. This kind of thing was bound to come biting our arses sooner or later. Getting the service running is not really complicated, but it does involve getting a domain and a wildcard cert which is a pain imho and not everyone can afford it so easily. I am currently struggling with getting a wildcard SSL properly into ACS as we speak. Realhostip.com should be kept running for the mid-term future, it will break a lot of stuff people run nowadays if it stops. Whoever owns the domain should donate it to someone who can and is willing to run it ... There's also the problem of the certificate. It comes bundled in ACS as far as I can tell.. When does it expire? What happens when it expires? 100 aleluias for removing this thing in 4.4! :) Lucian -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: [4.4] Feature request: root resize (FAO Marcus)
On 27.02.2014 09:29, Nux! wrote: Hi, Having talked about this issue recently, I'd like to request this feature so as to avoid template duplication. Marcus, you said you have a working patch, if you get it to apply to 4.4 I will invest time in testing it. I have opened an issue with rationale, example etc https://issues.apache.org/jira/browse/CLOUDSTACK-6181 Don't know what else I should be doing for a feature request. Lucian Anyone knows how to get Marcus' attention? -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
Re: Changing the SSL certificate for my own realhostip
On 28.02.2014 14:32, Wei ZHOU wrote: Can you visit the console by IE or Chrome? I managed to do it by modifying the DB, Billy's comments helped https://mail-archives.apache.org/mod_mbox/cloudstack-users/201402.mbox/%3C017b01cf348a%24c5739010%24505ab030%24%40dynamicquest.com%3E Also Chip's instructions helped clear the confusion a bit http://www.chipchilders.com/blog/2013/1/2/undocumented-feature-using-certificate-chains-in-cloudstack.html Thanks for help. We really need to get rid of this realhostip. :) -- Sent from the Delta quadrant using Borg technology! Nux! www.nux.ro
