Hi Drew, The log snippet you pasted doesn't really point to any actual error. It shows the planner starting to look for a destination host, however the reason it fails is probably further down in the logs, which would usually be along the lines of "No suitable hosts found" or host/pool being skipped for insufficient capacity. If you could share the rest of the management server logs specifically for log id: 64143c60, maybe we can help point you in the right direction. Also worth checking that your hosts have the capacity for this VM, which is requesting 16 GB RAM and ~99 GHz of CPU.
Regards. On Fri, Oct 23, 2026 at 12:40 PM Drew Weaver <[email protected]> wrote: > Hello, > > I did that prior to asking in the mailing list. > > They mostly express relief that it works when I explicitly tell it to put > the VM on a specific hypervisor and tell me to just do that. They almost > went as far to say that I am lucky "it works at all". > > Then if you push them further they complain about how the placement system > inside of Cloudstack is notoriously picky and hard to understand as a user > when it fails. > > Thank you, > -Drew > > > > > Op 22-10-2026 om 20:45 schreef Drew Weaver: > > Hello, > > > > Thanks you for making me aware of that. > > > > What I ended up doing was creating a new "Network Offering" and then > creating an additional new "Guest Network" using that new offering. > > > > If I create an instance in the CloudStack UI using this new Guest > Network (that is configured for ConfigDrive, etc) If I don't tell it > exactly what host to create the instance on during the creation process it > fails like this: > > > > 2026-10-22 13:22:29,719 DEBUG [c.c.d.DeploymentPlanningManagerImpl] > (API-Job-Executor-82:[ctx-ba6e2ef7, job-144, ctx-5ad14635]) > (logid:64143c60) Trying to deploy VM [error decoding VM instance > {"id":17,"instanceName":"i-2-17-VM","state":"Error","type":"User","uuid":"5eec0a30-a0e6-49a4-80d2-ba8aab230ddf"}] > and details: Plan > [{"_dcId":2,"_recreateDisks":false,"preferredHostIds":[],"migrationPlan":false,"hostPriorities":{}}]; > avoid list [{}] and planner: [null]. > > 2026-10-22 13:22:29,720 DEBUG [c.c.d.DeploymentPlanningManagerImpl] > (API-Job-Executor-82:[ctx-ba6e2ef7, job-144, ctx-5ad14635]) > (logid:64143c60) Checking non dedicated resources to deploy VM [VM instance > {"id":17,"instanceName":"i-2-17-VM","state":"Error","type":"User","uuid":"5eec0a30-a0e6-49a4-80d2-ba8aab230ddf"}]. > > 2026-10-22 13:22:29,723 DEBUG [c.c.d.DeploymentPlanningManagerImpl] > (API-Job-Executor-82:[ctx-ba6e2ef7, job-144, ctx-5ad14635]) > (logid:64143c60) Adding pods [[]], clusters [[]] and hosts [[]] to the > avoid list in the deploy process of user VM [error decoding VM instance > {"id":17,"instanceName":"i-2-17-VM","state":"Error","type":"User","uuid":"5eec0a30-a0e6-49a4-80d2-ba8aab230ddf"}], > because this VM is not explicitly dedicated to these components. > > 2026-10-22 13:22:29,723 DEBUG [c.c.d.DeploymentPlanningManagerImpl] > (API-Job-Executor-82:[ctx-ba6e2ef7, job-144, ctx-5ad14635]) > (logid:64143c60) Trying to allocate a host and storage pools from > datacenter [Zone {"id": "2", "name": "testberg", "uuid": > "31b0d190-f52c-44f9-9733-40e6e4d3a270"}], pod [null], cluster [null], to > deploy VM [VM instance > {"id":17,"instanceName":"i-2-17-VM","state":"Error","type":"User","uuid":"5eec0a30-a0e6-49a4-80d2-ba8aab230ddf"}] > with requested CPU [99200] and requested RAM [(16.00 GB) 17179869184]. > > 2026-10-22 13:22:29,723 DEBUG [c.c.d.DeploymentPlanningManagerImpl] > (API-Job-Executor-82:[ctx-ba6e2ef7, job-144, ctx-5ad14635]) > (logid:64143c60) ROOT volume [null] is not ready to deploy VM [VM instance > {"id":17,"instanceName":"i-2-17-VM","state":"Error","type":"User","uuid":"5eec0a30-a0e6-49a4-80d2-ba8aab230ddf"}]. > > 2026-10-22 13:22:29,727 DEBUG [c.c.d.DeploymentPlanningManagerImpl] > (API-Job-Executor-82:[ctx-ba6e2ef7, job-144, ctx-5ad14635]) > (logid:64143c60) DeploymentPlan [DataCenterDeployment] has not specified > host. Trying to find another destination to deploy VM [VM instance > {"id":17,"instanceName":"i-2-17-VM","state":"Error","type":"User","uuid":"5eec0a30-a0e6-49a4-80d2-ba8aab230ddf"}], > avoiding pods [], clusters [] and hosts []. > > > > If I tell it exactly where to place the VM in the UI [there is only one > zone, one pod, one cluster, and one host] it can create the VM... > > > > When looking at the two Guest Network configurations they are basically > the same except for the things that have to be unique (IP Addresses, vlan > ID, etc) and the Network Offering that is associated to each one. > > > > The Network offerings are basically the same as well except for the new > one has 5000 set for the speed (instead of 200) and DHCP, DNS, and UserData > are all set to "ConfigDrive". > > > > Thank you for any guidance that anyone can offer. > > What helps here, use Claude Code or OpenAI Codex, tell them you are using > Apache CloudStack version XX and throw in the loglines. It will most of the > time exactly tell you what is going wrong. > > Wido > > > -Drew > > > > > > > > -----Original Message----- > > From: Pearl Dsilva <[email protected]> > > Sent: Thursday, October 22, 2026 12:20 PM > > To: [email protected] > > Subject: Re: Network offering => Network, question > > > > Hi Drew, > > > > With respect to the issue you're noticing where clone offering silently > discards changes made to an offering's service list, seems to be a known UI > issue, and has a fix: > https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_apache_cloudstack_pull_14062&d=DwIFaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=OPufM5oSy-PFpzfoijO_w76wskMALE1o4LtA3tMGmuw&m=0zXGfDiufAOL6uejZHU0ecbHAZVz5s2n6XnPApFNJlcRfoeBqi0D2XSsySx0MG0W&s=t1dSViNUtWHcwcj1sv6camXYUkCAGVVAaBgC-MYktIg&e= > . > > Until this is merged, as a workaround, you may have to create a new > offering with your desired list of services. > > > > Regards, > > Pearl > > > > On Thu, Oct 22, 2026 at 11:51 AM Drew Weaver <[email protected]> > wrote: > > > >> Hello, > >> > >> I've made some progress setting up Cloudstack (thanks everybody that > >> responded to my questions). > >> > >> I noticed on the test VM I spun up that it's using DHCP (without a > >> default > >> gateway) and while it can reach the Internet (public IP addresses) > >> using DHCP isn't desirable for our goals. > >> > >> Also it isn't immediately obvious how the VM is routing to the > >> Internet without a gateway configured in the network settings since > >> there is no NAT of any kind taking place. > >> > >> So I cloned "DefaultSharedNetworkOffering< > >> > https://urldefense.proofpoint.com/v2/url?u=http-3A__10.1.225.2-3A8080_client_-23_networkoffering_9fc257f7-2D62cb-2D476a-2Db3e3-2Ddca3451f4eb1&d=DwIFaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=OPufM5oSy-PFpzfoijO_w76wskMALE1o4LtA3tMGmuw&m=0zXGfDiufAOL6uejZHU0ecbHAZVz5s2n6XnPApFNJlcRfoeBqi0D2XSsySx0MG0W&s=GQJMYv_F7vrpEQSiuOFS5pxW2FfNxshGhZBgvr73Ktw&e= > >" > >> and changed DHCP, DNS, and UserData to "ConfigDrive" from DHCP and > >> then configured "SecurityGroups" as "SecurityGroupProvider". > >> > >> I am guessing that I did something wrong because I noticed when I > >> saved it, it kind of discarded everything I changed and made the new > >> offering like this: > >> > >> Supported services > >> Dhcp : VirtualRouter > >> UserData : VirtualRouter > >> Dns : VirtualRouter > >> SecurityGroup : SecurityGroupProvider > >> > >> Does anyone have any insight into how I can configure this so it > >> doesn't use virtual routers while preventing IP spoofing/nic stepping? > >> > >> Thanks, > >> -Drew > >> > >> > >> > >> > >> > > > > -- > > Pearl Dsilva > > -- Pearl Dsilva
