Re: Unable to ping System VM in Advanced Zone (without SG)
Have you configured the vlan tag of public ip range ? -Wei On Friday, 16 June 2023, Pratik Chandrakar wrote: > Hi, > I am pinging the Public IP. I tried to ping the gateway from CPVM also but > was unable to connect to the gateway, whereas in the SG-enabled setup, the > same IP and configuration work. I have checked the switch as well and I am > getting the MAC of the System VMs. > Unable to identify this strange issue. > > > On Fri, Jun 16, 2023 at 11:48 AM Wei ZHOU wrote: > > > Hi, > > > > System vms have multiple IPs, which IP do you ping to? > > > > -Wei > > > > On Thursday, 15 June 2023, Pratik Chandrakar > > > wrote: > > > > > Hi Wei, > > > Switch ports are in trunk mode. > > > VM on the same host is also not working even though its agent state is > > > showing up. > > > I think the Configuration is also correct because the same > configuration > > is > > > working for SG-enabled Zone. > > > > > > On Thu, Jun 15, 2023 at 3:30 PM Wei ZHOU > wrote: > > > > > > > Hi, > > > > > > > > If VMs on the same host work, but not on different hosts - this is a > > > > typical issue of network misconfiguration. > > > > Please check if switch ports are trunk mode, hosts and cloudstack > > network > > > > configurations are correct. > > > > > > > > -Wei > > > > > > > > On Thu, 15 Jun 2023 at 11:46, Pratik Chandrakar < > > > > chandrakarpra...@gmail.com> > > > > wrote: > > > > > > > > > Hi all, > > > > > In the Advanced Zone-SG Enabled deployment there is no issue but on > > the > > > > > same setup when I deploy Advanced Zone-Without SG, I am not able to > > > ping > > > > > System VMs from other client machines/CS hosts. Both Console Proxy > > and > > > > > Secondary storage are UP and agent states are also UP in case > > > Management > > > > > Server and System VMs are deployed on the same host whereas when > > System > > > > VMs > > > > > are deployed on other hosts then it does not show any state in > > > Management > > > > > Server. I checked the logs but couldn't find any error on both > > > management > > > > > and agent logs. > > > > > Environment Details are > > > > > ACS - 4.18 > > > > > Host - Ubuntu 22.04.2 > > > > > > > > > > Please guide me as I am unable to find any solution to this strange > > > > issue. > > > > > > > > > > -- > > > > > *Regards,* > > > > > *Pratik Chandrakar* > > > > > > > > > > > > > > > > > > -- > > > *Regards,* > > > *Pratik Chandrakar* > > > > > > > > -- > *Regards,* > *Pratik Chandrakar* >
Re: Unable to ping System VM in Advanced Zone (without SG)
Hi, I am pinging the Public IP. I tried to ping the gateway from CPVM also but was unable to connect to the gateway, whereas in the SG-enabled setup, the same IP and configuration work. I have checked the switch as well and I am getting the MAC of the System VMs. Unable to identify this strange issue. On Fri, Jun 16, 2023 at 11:48 AM Wei ZHOU wrote: > Hi, > > System vms have multiple IPs, which IP do you ping to? > > -Wei > > On Thursday, 15 June 2023, Pratik Chandrakar > wrote: > > > Hi Wei, > > Switch ports are in trunk mode. > > VM on the same host is also not working even though its agent state is > > showing up. > > I think the Configuration is also correct because the same configuration > is > > working for SG-enabled Zone. > > > > On Thu, Jun 15, 2023 at 3:30 PM Wei ZHOU wrote: > > > > > Hi, > > > > > > If VMs on the same host work, but not on different hosts - this is a > > > typical issue of network misconfiguration. > > > Please check if switch ports are trunk mode, hosts and cloudstack > network > > > configurations are correct. > > > > > > -Wei > > > > > > On Thu, 15 Jun 2023 at 11:46, Pratik Chandrakar < > > > chandrakarpra...@gmail.com> > > > wrote: > > > > > > > Hi all, > > > > In the Advanced Zone-SG Enabled deployment there is no issue but on > the > > > > same setup when I deploy Advanced Zone-Without SG, I am not able to > > ping > > > > System VMs from other client machines/CS hosts. Both Console Proxy > and > > > > Secondary storage are UP and agent states are also UP in case > > Management > > > > Server and System VMs are deployed on the same host whereas when > System > > > VMs > > > > are deployed on other hosts then it does not show any state in > > Management > > > > Server. I checked the logs but couldn't find any error on both > > management > > > > and agent logs. > > > > Environment Details are > > > > ACS - 4.18 > > > > Host - Ubuntu 22.04.2 > > > > > > > > Please guide me as I am unable to find any solution to this strange > > > issue. > > > > > > > > -- > > > > *Regards,* > > > > *Pratik Chandrakar* > > > > > > > > > > > > > -- > > *Regards,* > > *Pratik Chandrakar* > > > -- *Regards,* *Pratik Chandrakar*
Re: Direct Download/Bypass Secondary Storage option for templates
Nicolas, "In your requirement do you mean to change the template URL and re-download the template from a different location or simply download again the template from the same URL after updating the served file?" For our use case the URL will stay the same, but the file will be updated (though I could imagine other use cases where an update of the URL would be useful too). Consider a URL like http://some.distro.com/cloud/latest/imgname.img where /latest/ automatically takes you to the latest updated version of that file over time. We would not be looking to trigger an update on all of the primary storage pools where instances have been deployed, just the secondary storage such that new deploys use the updated image. Over time, distro cloud images get updated with the latest updates (that one would receive just by running apt update && apt upgrade, for instance). If we continue to deploy with the same image, updates will cause vm deployment to take longer and longer over time, as well as consume more and more network bandwidth, due to the updates each vm will have to perform at time of deployment. "Afaik there is no supported way to safely do this except manually updating the files on the downloaded storage pools (secondary and primary) which should be updated carefully. The tables template_store_ref and template_spool_ref indicate the downloaded copies on secondary and primary pools of each template." I assume you're referring to the question about copying over/updating the secondary storage image manually. I consider this idea a hack that could cause issues or at the very least be at risk of breaking when a new cloudstack update gets deployed. I don't like the idea myself. I'm just trying to explore all the options available to us and see where your guidance takes me. The reality is that we can make direct download work, but to do so would require additional infrastructure (for instance, a webserver in each zone where we point the URL to, that we now have to maintain and monitor). Ideally we would just like to let the secondary storage server(s) fulfill their role. On Thu, Jun 15, 2023 at 10:30 PM Nicolas Vazquez < nicolas.vazq...@shapeblue.com> wrote: > Hi Will, > > When registering a template CloudStack usually downloads it first to > secondary storage and then copies it through the different primary storage > pools when a VM deployment requires it. The aim of the direct download > feature is to skip the first step and directly downloading the templates > into primary storage pools without any secondary storage intervention. > > In your requirement do you mean to change the template URL and re-download > the template from a different location or simply download again the > template from the same URL after updating the served file? Afaik there is > no supported way to safely do this except manually updating the files on > the downloaded storage pools (secondary and primary) which should be > updated carefully. The tables template_store_ref and template_spool_ref > indicate the downloaded copies on secondary and primary pools of each > template. > > Regards, > Nicolas Vazquez > > > From: Will Conrad > Date: Thursday, 15 June 2023 at 14:47 > To: users@cloudstack.apache.org > Subject: Re: Direct Download/Bypass Secondary Storage option for templates > Nicolas, > > The reason we're considering using the directdownload feature is to > simplify template maintenance/updates. I presume that's what it was > designed for. We want to be able to, preferably through cloudstack > functionality, update the template image file associated with a template. > We planned on achieving this utilizing directdownload to decouple the image > file from the registered template itself when it occurred to us that a > "regrab" button in the template properties webui or an API call to tell > secondary storage to redownload the source would very much simplify this > process. > > This brings my questions to: > > How difficult would it be to implement something like that? > > Is there another way to update the imagefile associated with a template? I > mean, could we manually overwrite the file on secondary storage? Would that > break anything? > > What is Cloudstack's recommended best practice for managing template > images? > > > Regards, > > Willard (Will) > > > > On Wed, Jun 14, 2023 at 10:26 AM Nicolas Vazquez < > nicolas.vazq...@shapeblue.com> wrote: > > > No problem, I think these docs do not clearly state the supported storage > > providers, I will fix that. On this blog entry we have mentioned them: > > > https://www.shapeblue.com/cloudstack-feature-first-look-direct-download-agnostic-of-the-storage-provider/ > > > > Currently the direct download feature is supported on NFS, local storage > > and shared mount point, but not for Ceph. > > > > Regards, > > Nicolas Vazquez > > > > > > From: Will Conrad > > Date: Wednesday, 14 June 2023 at 10:58 > > To: users@cloudstack.apache.org > > Subject: Re: Direct Download/Bypass
RE: Restarting multiple VPCs simultaneously
Hi Wei, You're right, I was able to restart the VPCs simultaneously. Was just using the wrong command -_- Thanks -Original Message- From: Wei ZHOU Sent: Thursday, June 15, 2023 5:56 PM To: users@cloudstack.apache.org Subject: Re: Restarting multiple VPCs simultaneously Hi Frederic, I do not think there is such restriction in cloudstack ? Have you tried it on UI ? -Wei On Thursday, 15 June 2023, Frederic Larcher wrote: > Hi, > > I am using Cloudstack 4.15.2. > > Is there a way to force Cloudstack to restart multiple VPCs at once? > From what I have seen so far, Cloudstack won't restart a VPC while the > previous one has not finished being recreated, so it is taking quite a > while if I need to restart all the VPCs. > > I guess it is by design, and it consumes a good amount of resources, > but I wanted to check if there was a way to speed up the process? > > Thanks >
TAC Applications for Community Over Code North America and Asia now open
Hi All, (This email goes out to all our user and dev project mailing lists, so you may receive this email more than once.) The Travel Assistance Committee has opened up applications to help get people to the following events: *Community Over Code Asia 2023 - * *August 18th to August 20th in Beijing , China* Applications for this event closes on the 6th July so time is short, please apply as soon as possible. TAC is prioritising applications from the Asia and Oceania regions. More details on this event can be found at: https://apachecon.com/acasia2023/ More information on how to apply please read: https://tac.apache.org/ *Community Over Code North America - * *October 7th to October 10th in Halifax, Canada* Applications for this event closes on the 22nd July. We expect many applications so please do apply as soon as you can. TAC is prioritising applications from the North and South America regions. More details on this event can be found at: https://communityovercode.org/ More information on how to apply please read: https://tac.apache.org/ *Have you applied to be a Speaker?* If you have applied or intend to apply as a Speaker at either of these events, and think you may require assistance for Travel and/or Accommodation - TAC advises that you do not wait until you have been notified of your speaker status and to apply early. Should you not be accepted as a speaker and still wish to attend you can amend you application to include Conference fees, or, you may withdraw your application. The call for presentations for Halifax is here: https://communityovercode.org/call-for-presentations/ and you have until the 13th of July to apply. The call for presentations for Beijing is here: https://apachecon.com/acasia2023/cfp.html and you have until the 18th June to apply. *IMPORTANT Note on Visas:* It is important that you apply for a Visa as soon as possible - do not wait until you know if you have been accepted for Travel Assistance or not, as due to current wait times for Interviews in some Countries, waiting that long may be too late, so please do apply for a Visa right away. Contact tac-ap...@tac.apache.org if you need any more information or assistance in this area. *Spread the Word!!* TAC encourages you to spread the word about Travel Assistance to get to these events, so feel free to repost as you see fit on Social Media, at work, schools, universities etc etc... Thank You and hope to see you all soon Gavin McDonald on behalf of the ASF Travel Assistance Committee.
Re: Unable to ping System VM in Advanced Zone (without SG)
Hi, System vms have multiple IPs, which IP do you ping to? -Wei On Thursday, 15 June 2023, Pratik Chandrakar wrote: > Hi Wei, > Switch ports are in trunk mode. > VM on the same host is also not working even though its agent state is > showing up. > I think the Configuration is also correct because the same configuration is > working for SG-enabled Zone. > > On Thu, Jun 15, 2023 at 3:30 PM Wei ZHOU wrote: > > > Hi, > > > > If VMs on the same host work, but not on different hosts - this is a > > typical issue of network misconfiguration. > > Please check if switch ports are trunk mode, hosts and cloudstack network > > configurations are correct. > > > > -Wei > > > > On Thu, 15 Jun 2023 at 11:46, Pratik Chandrakar < > > chandrakarpra...@gmail.com> > > wrote: > > > > > Hi all, > > > In the Advanced Zone-SG Enabled deployment there is no issue but on the > > > same setup when I deploy Advanced Zone-Without SG, I am not able to > ping > > > System VMs from other client machines/CS hosts. Both Console Proxy and > > > Secondary storage are UP and agent states are also UP in case > Management > > > Server and System VMs are deployed on the same host whereas when System > > VMs > > > are deployed on other hosts then it does not show any state in > Management > > > Server. I checked the logs but couldn't find any error on both > management > > > and agent logs. > > > Environment Details are > > > ACS - 4.18 > > > Host - Ubuntu 22.04.2 > > > > > > Please guide me as I am unable to find any solution to this strange > > issue. > > > > > > -- > > > *Regards,* > > > *Pratik Chandrakar* > > > > > > > > -- > *Regards,* > *Pratik Chandrakar* >
