Hi Amit,
Ingress and Egress rules only have an effect when the traffic goes through the
virtual router. Conversely, DHCP traffic is normally not meant to pass through
routers or gateways. But from your description, I gather you want a shared
network anyway.
Did you run a tcpdump on the host t
Hi Andrija,
My idea would be to either ensure (in the cloudstack-setup-management) that
both firewalld/ufw are disabled and continue operating with pure iptables
OR to not add rules at all, but instead print a message on the
requirements to open access to ports 8080/8250/9090 with whatever firew
so there is not enough data.
But I manually started a VM with 32G memory on KVM HOST, and it succeeded; and
Buffers/cached/memfree/they were reduced accordingly.
therefore:
Can we add the memory occupied by Buffers to AvailableMemory and report to
CLOUDSTACK?
-邮件原件-
发件人: Riepl, Gre
ce/LibvirtVMDef.java#L980>
Regards,
> On 15 Jul 2020, at 12:00, Riepl, Gregor (SWISS TXT)
> wrote:
>
> Hi El Nino,
>
> Wow, you are actually correct.
> QEMU (the hardware emulation layer KVM uses) does have vmxnet3 emulation:
> https://wiki.qemu.org/ChangeLog/1.5#Network_car
o-bus
name "virtio-net-pci", bus PCI, alias "virtio-net"
name "vmxnet3", bus PCI, desc "VMWare Paravirtualized Ethernet v3”
After deploying the instance from cloudstack GUI, and do a virsh edit on the
kvm host to change e1000 by vmxnet3 and reboot the ins
This sounds very familiar...
https://bugzilla.redhat.com/show_bug.cgi?id=1250060
and they rolled it back in
https://bugzilla.redhat.com/show_bug.cgi?id=1779609
because upstream didn't accept their patch.
We had to "fix" our memory monitoring twice because of this.
It's important to *not* calcula
+1 (non-binding)
I tested a few common and less-common things and found no regressions.
Note: I built the binary locally, didn't try the binaries on the Github release
page.
From: Rohit Yadav
Sent: 01 July 2020 06:51
To: d...@cloudstack.apache.org ;
users@clouds
Hi El Nino,
I think there is a bit of a misconception here: vmxnet3 is a specific,
proprietary, paravirtualized network interface provided by VMware (ESX).
There is no vmxnet3 support in KVM. KVM provides its own paravirtualized
network interface, called virtio_network:
https://www.linux-kvm.o
Seems to be this line:
https://github.com/apache/cloudstack/blob/4.13.1.0/server/src/main/java/com/cloud/resource/ResourceManagerImpl.java#L1742
Judging from the surrounding code, the NPE happens at the .getName() call.
Could it be that you haven't got a datacenter assigned to the pod?
__
Hi gu haven,
There is also the UserData service that can be enabled in the virtual router
(actually: on the Network Offering).
It provides simple HTTP-based access to the metadata.
Access Control is done through the host's IP or MAC address, as far as I know.
https://docs.cloudstack.apache.org/en
not 100% certain this is exactly what Bobby was saying,
> but if this is the case then I think it should not block us.
>
> On Wed, May 20, 2020 at 1:00 AM Riepl, Gregor (SWISS TXT) <
> gregor.ri...@swisstxt.ch> wrote:
>
> > Hi everyone
> >
> > Sorry for the late
Hi everyone
Sorry for the late response, but I have a few concerns:
* As Bobby stated, this bug seems to only occur with VMware 6.7+, and it
sounds to me like they should take action on it. Does someone track this with
VMware?
* Do I understand correctly that the issue only occurs when
Hi Rohit,
Let me comment on just a few of the topics:
> Release cycle
I think we should definitely have a daily/nightly build, at least as long as a
lot of changes are incoming.
I'm think along the lines of a parallel installation, so all versions can be
tested and users still get a fallback
Previously, the common way to access the system VMs was via SSH from the
management server:
https://docs.cloudstack.apache.org/projects/archived-cloudstack-getting-started/en/latest/administration_guide.html#accessing-system-vms
It looks like this piece of documentation is missing from the newer
Sent: 04 May 2020 11:35
To: users
Subject: Re: [VOTE] Apache CloudStack 4.14.0.0 RC1
Gregor,
is this something you've also hit during testing 4.14 only, or are you
aware of the need for setting those values even before (4.11-4.13) ?
Thanks
Andrija
On Mon, 4 May 2020 at 10:49, Riepl, Gregor
Hi Liridon,
Note that
> SET GLOBAL time_zone = '+2:00';
has the mostly same effect as writing
> default-time-zone= "-05:00"
to /etc/my.cnf
The difference is that using SET GLOBAL does not persist the setting across
MySQL starts. You should write this setting into the configuration file to ma
Hi Olivier,
It may also be a good idea to check your network offerings, perhaps there is an
incorrect limit somewhere that wasn't applied previously and is now limiting
the bandwith.
I remember having such an issue after an upgrade once.
Regards,
Gregor
From: Vi
Hi Vivek + Fariborz,
Are you sure this is the issue here?
The Python connector should be very much backwards-compatible with any older
MySQL version.
We're using the most recent version with a 5.x MySQL ourselves, and it seems
upstream has removed the older packages from their package server any
Hi Christian,
The feature you mention is not in vanilla CloudStack yet, at least not that I
know of.
See the discussion here for more info:
https://lists.apache.org/thread.html/feccbf1a6d7e8756bb0d799653e325c929356d71ff2227bce0a1b9a4%40%3Cdev.cloudstack.apache.org%3E
@ustcweizhou Did you prepar
Hi Jagdish,
I'm not sure why your pods aren't able to communicate with each other.
The network interfaces that ACS sets up are linked through a L2 virtual switch
and should be able to talk to each other directly.
Which networking service are you using?
https://kubernetes.io/docs/concepts/cluste
Hi Jerry,
I think this is because of the way a VM is tied to its root disk, and how a
root disk is chained to its template.
VMs created from ISO do not have such a chain.
Technically, it shouldn't be an issue to swap out the root disk entirely,
though...
Regards,
Gregor
___
Thanks for the clarification, Andrija.
We're mostly using ESX here, but I do think it will work similarly, with the
limit being applied to the vNIC.
From: Andrija Panic
Sent: 22 November 2019 16:32
To: Riepl, Gregor (SWISS TXT)
Cc: users
Subject: Re:
Is this really applied to each VM?
I was under the impression that you could only limit network bandwidth on the
network offering.
How is this implemented exactly? Does it limit the NIC speed of the individual
VMs, or does it only apply to external traffic on the VR?
___
Hi Fariborz,
Sorry, I don't quite understand what you're referring to.
For Advanced Networking with a virtual router, you have to create egress rules
yourself, using
https://cloudstack.apache.org/api/apidocs-4.11/apis/createEgressFirewallRule.html
or the UI. The same applies to VPCs.
On Basic
Hi Steve,
We won't be able to attend the event, unfortunately, but the topics look very
interesting.
Will there be recordings and/or live streams of the sessions?
Regards,
Gregor
From: Steve Roles
Sent: 07 October 2019 18:02
To: users@cloudstack.apache.org ;
ma
Hi CloudStack community,
We're still stuck with cloud-init's problematic way of searching for a metadata
server: https://github.com/apache/cloudstack/issues/3557
Would you consider taking a look at our proposed fix and comment on it, so it
can be pushed ahead?
https://code.launchpad.net/~joschi
Hi Rohit
+1
Kudos to the plans and the work already done on the new UI!
I'm definitely in favour of something more modern and easier to maintain.
Also, the complete separation of the UI from the mgmt server is definitely the
way to go.
We've tested the current state of the frontend and while a
> When I pass base64 encoded user data to deployvirtualmachine Api, VR
> returns empty script upon calling http://VR_IP/latest/user-data from
> within
> the VM. ACS version is 4.12.0.0. It was working before. I've tried
> restarting network.
What do you mean exactly, when you say "It was working
Thanks, everybody!
From: Rohit Yadav
Sent: 02 September 2019 09:41
To: d...@cloudstack.apache.org ;
users@cloudstack.apache.org
Subject: Re: [VOTE][RESULT] Apache CloudStack 4.13.0.0
Great, congrats all.
Regards,
Rohit Yadav
Software Architect, ShapeBlue
ht
Rohit answered this:
https://mail-archives.apache.org/mod_mbox/cloudstack-dev/201905.mbox/%3cvi1pr07mb5117016e3ec7b3e523aa1f7ae9...@vi1pr07mb5117.eurprd07.prod.outlook.com%3e
Disabling the unit tests seems to be easiest solution for now.
A fix would probably involve a proper TLS setup for this u
Hi Fariborz,
I would consider both ways to monitor the instance.
Even if the hypervisor can give you accurate information about allotted
CPU time and memory, there are always some things that it doesn't see.
As for monitoring agents, it highly depends on what monitoring system
you use or want to
Hi Nick,
This might not be relevant for Xen, but we've had problems with memory leaks on
the VRs on VMware when balloon memory was enabled.
A while ago, we built a custom router monitoring setup via SSH for our
environment, because CloudStack doesn't give us enough information about router
sta
You can change the async_job DB record to error or similar, but that will most
likely not stop the task.
We've done this in the past, but only when a job was stuck forever due to
incorrect error handling and not when it was still running.
Is there nothing in the API to stop a running task?
and here:
https://www.shapeblue.com/dynamic-roles-in-cloudstack/
more fun facts:
http://events17.linuxfoundation.org/sites/events/files/slides/CCCNA17%20CS%20Dynamic%20Roles%20in%20Cloudstack.pdf
Hope this helps!
Regards,
Gregor
On Wed, 2019-07-17 at 11:48 +, Riepl, Gregor (SWISS TXT) wrote
Hi Richard,
By default, CloudStack points the remote console to int-ern-al-
ip.your.domain. As a quick fix, you need to point this domain to the
external IP under which the console proxy is reachable.
I think it's also possible to customise the domain name, so you don't
have to add a new DNS reco
Congratulations, Nathan!
From: Paul Angus
Sent: 19 July 2019 16:58
To: d...@cloudstack.apache.org ;
users@cloudstack.apache.org ;
priv...@cloudstack.apache.org
Subject: [ANNOUNCE] Nathan Johnson has joined the PMC
Fellow CloudStackers,
It gives me great plea
peblue.com
____
From: Riepl, Gregor (SWISS TXT)
Sent: Thursday, July 11, 2019 3:15:12 PM
To: priv...@cloudstack.apache.org ;
d...@cloudstack.apache.org
Cc: users@cloudstack.apache.org
Subject: Re: CloudStack Kubernetes Provider
Hi Rohit,
> One of
Hi Eiji,
> mysql> select * from cloud.account where id = 2;
> ++--+--+
> --+---+-+-++--
> --+-+-+
> | id | account_name | uuid |
Hi Jean-François
> I have always used NFS to install the SSVM templates and the install
> script (cloud-install-sys-tmplt) only takes a mount point. How, if
> possible, would I proceed with S3 only storage ?
CloudStack doesn't support object storage as a backend for the
secondary storage. You'd
;_=1561961591972 HTTP/1.1" 432 151
> > >
> > > This happens just after login and indicates something is wrong.
> > > It might not help much, but can you post the JSON body of the
> > > response?
> >
> > This JSON body was returned.
Congratulations, Boris!
Congratulations, everyone!
Good work everybody!
This release contains a few crucial fixes for us, so we're happy it's
there!
On Sat, 2019-07-13 at 16:52 +0100, Paul Angus wrote:
> Announcing Apache CloudStack LTS Maintenance Release 4.11.30
>
>
>
> The Apache CloudStack project is pleased to announce the release of
> Cl
Hi Rohit,
> One of the community contributors from SwissTxt Gregor Riepl
> (@onitake) have also offered to contribute their provider (which is
> already under the Apache v2.0 license) which they have based on the
> original provider:
> https://github.com/kubernetes/enhancements/issues/672#issueco
Hi Eiji,
> 192.168.122.1 - - [01/Jul/2019:06:09:22 +0900] "GET
> /client/api?command=listCapabilities&response=json&sessionkey=null&_=
> 1561961364158 HTTP/1.1" 401 101
This happens before login and is ok.
I see that too when I open the login page without an active session.
> ==> /var/lo
> https://github.com/apache/cloudstack/pull/3417
>
>
> The issue only affects 4.11.2.0 users for which the upgrade path does
> not run (a noop upgrade path runs) and it does not automatically fix
> any registered 4.11.3 systemvmtemplate to be used post-upgrade. I've
> mentioned workaround steps
> @Riepl
> nmap -sS -O does help in fetching the OS type only if they have
> public ip. I cant ssh into the machines because they are customer
> machines and I dont have credentials for them.
We had such a situation a few times, and simply asked the affected
customer if they would permit us to de
> version. Another way is to open the console and see the login screen.
> This will get the actual data but I want to do automation to see for
> all VM's and opening the console is not feasible to automate. Is
> there any other way to get it?
Are the VMs networked?
You could fetch their public I
+1
Based on:
- Packages from http://packages.shapeblue.com/testing/41130rc1/ and
http://packages.shapeblue.com/testing/systemvm/41130rc1/
- Upgrade of our test cloud from 4.11.2 to 4.11.3 worked without problems
(aside from the mentioned template issue)
- Our internal smoke test stack ran succes
s DB values so it becomes a "SYSTEM"
instead of "USER" type that it was initially.
Andrija
On Wed, 12 Jun 2019 at 16:46, Riepl, Gregor (SWISS TXT) <
gregor.ri...@swisstxt.ch> wrote:
> Hi Paul,
>
> > The release notes are still work-in-progress, but the
;s DB values so it becomes a "SYSTEM"
instead of "USER" type that it was initially.
Andrija
On Wed, 12 Jun 2019 at 16:46, Riepl, Gregor (SWISS TXT) <
gregor.ri...@swisstxt.ch> wrote:
> Hi Paul,
>
> > The release notes are still work-in-progress, but the s
Hi Paul,
> The release notes are still work-in-progress, but the systemvm
> template upgrade section has been updated. You may refer the
> following for systemvm template upgrade testing:
> http://docs.cloudstack.apache.org/projects/cloudstack-release-notes/en/latest/index.html
Thanks!
We're doi
> Usually indicates missing storage driver in initramfs (if you were on
> VMware, I would tell you to check if vmw_pvscsi.ko is present in
> initramfs).
First thing to check would be to compare the list of PCI devices
(lspci) and storage devices (cat /proc/partitions) between the older
and the ne
Hi Vladimir
The CloudStack cloud provider code will be removed from the k8s core
soon: https://github.com/kubernetes/kubernetes/pull/72178
It's not advisable to use it any more.
You can use our CCM that is based on the old cloud provider instead:
https://github.com/swisstxt/cloudstack-cloud-cont
+1
Thanks René for all your hard work and the good times we had when we worked on
the same team.
I'll still be sticking around, but likely won't be as involved and active as
René was.
There's too many other things needing my attention. ;)
From: Rene Moser
Sent:
Controller type as you need it - this will make sure all new VMs
deployed from that template will inherit the specific root controller
type...
Let me know if that works for you.
Andrija
On Fri, 17 May 2019 at 15:07, Riepl, Gregor (SWISS TXT) <
gregor.ri...@swisstxt.ch> wrote:
> Hi Andrij
t controller driver in
Windows can sometimes cause it to fail boot.
Best,
Andrija
On Thu, 16 May 2019 at 21:44, Riepl, Gregor (SWISS TXT) <
gregor.ri...@swisstxt.ch> wrote:
>
> > $ cloudstack addResourceDetail "details[0].key=rootDiskController"
> > "details[0
> $ cloudstack addResourceDetail "details[0].key=rootDiskController"
> "details[0].value=lsisas1068" "resourcetype=UserVm" "resourceid=$id"
Note: lsisas1068 comes from
https://github.com/apache/cloudstack/blob/4.11.2.0/vmware-base/src/com/cloud/hypervisor/vmware/mo/ScsiDiskControllerType.java
W
Hi,
This is an old topic, [1] but it looks like behaviour changed in newer
CloudStack versions.
What is the recommended procedure to change the NIC and disk controller
types of an existing VM (in stopped state)?
We used
$ cloudstack addResourceDetail "details[0].key=rootDiskController"
"details
> What we'd like to know is realtime stats about the number of NAT/TCP
> sessions (i.e. conntrack status), CPU load, memory/swap usage,
> network
> and I/O bandwidth, and possibly some process metrics.
In addition to that, there's more checks that would make sense (and are
difficult or impossible
> This include stats from virtual router as well since libvirt exporter
> use virsh libraries to fetch details. For libvirt, virual router is
> just another VM and so it will exporter those stats also.
>
> Run virsh list and you will see virual router entry and using
> dommemstats or other comman
> You can use "libvirt exporter" to export all stats like cpu utilization,
> memory consumption, disk read/write, network read/write and export it to
> prometheus. From prometheus you can use grafana or other monitoring tools
> to see all stats. We use exporter merger third party tool to merge all
> How I got stats for all VMs, including system ones, was to install
> Prometheus, both the cloudstack and the libvirt exporter.
> I know it's "outside" Cloudstack, but may prove useful.
It looks like the Prometheus exporter is limited to per-zone statistics, though?
https://cwiki.apache.org/conf
Hi Boris,
> I don’t think you could directly query for a VR, but you could specify a
> network ID and it’ll list you all the metrics of VMs on that network. Then I
> suppose you could work out a summary from that list.
> https://cloudstack.apache.org/api/apidocs-4.12/apis/listVirtualMachinesMetr
Hi,
I recently discovered that CloudStack exposes certain VM metrics like disk I/O,
network bandwidth, CPU load and (if the HV agent is installed) memory usage.
Presumably, these are fetched directly from the hypervisor.
However: I found no way to get the same statistics for system VMs, in
part
> You should be using a computer offering with HA enabled.
> When you’re using this option, cloudstack will start vm until you shutdown it
> by cloudstack interface (UI or api).
I think you meant: You should NOT be using a compute offering with HA enabled,
right?
> I have a problem with my cloud-management web UI. It just stopped
> accepting connections and gives apache error.
>
> Is theer anyway I could get some help from you?
Is this related to Jevgeni's issue or a completely new one?
You shouldn't simply hit reply on a mail thread, as that will make t
Hi Jevgeni,
> (1) client -> VM1:80/app -> VM2:8080/app
> (2) client -> VM1:80/data -> VM3:8080/data
>
> This was working fine before the reinstallation.
> We found that it works, if we stop iptables.
>
> But with iptables ON, (1) works, but (2) does not work - it gives
> connection refused.
> Ho
> I create a new iconset for cloudstack and we think we will implement
> it. There is a Vote Formular and some examples. I oriented myself on
> the original one. Its a lot of work to make them all looks good but
> if you do not start it will be nothing. Lets start with the dashboard
> and left con
No sure if my vote counts, but a
[x] +1 approve
from me.
I built the tag 6.0.0 locally with Go 1.11.5 and tested against a CS
4.5.2 and a CS 4.11.2 instance.
Listing VMs and stopping/starting instances works as expected, on both
CloudStack versions.
I also filed a few minor issues on Github, but
> I’m happy to announce that the CloudStack PMC has elected Paul Angus
> as our new VP of CloudStack.
Congratulations, Paul!
> If you can, would you mind submitting a pull request to the
> documentation repo
> https://github.com/apache/cloudstack-documentation
> to add the missing steps?
Done: https://github.com/apache/cloudstack-documentation/pull/30
> If you can, would you mind submitting a pull request to the
> documentation repo
> https://github.com/apache/cloudstack-documentation
> to add the missing steps?
Sure!
I should note that I haven't tested the exact steps on a custom CS
repository. The script I posted is one that I use for a per
Hi,
> I'm not sure the community is able to help you with local DEB repo
> you host
> in the network. Try to use the guide and install from the official
> repository.
I think Alejandro *was* following the guide. It specifically explains
how to build your own CloudStack packages and set up a dpkg
Hi Ivan,
> The question is how to cut DNS feature for a running Zone (Basic Zone
> With
> SG) safely without stopping the operation. Anybody has dealt with
> that?
> What is your suggestions about possible approaches.
How about ignoring the nameservers supplied via DHCP?
For example, if you use
Hi all,
with the more-or-less recent decision to factor all cloud-controller-
managers out of the Kubernetes core [1], we decided to fork the
existing code into a new project: [2]
As written here [3], the migration to standalone CCMs is still ongoing
and marked as beta in Kubernetes 1.13.
This C
77 matches
Mail list logo