You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cloudstack.apache.org by "weizhouapache (via GitHub)" <gi...@apache.org> on 2026/10/15 11:01:05 UTC
[PR] Main fix 13484 14321 [cloudstack]
weizhouapache opened a new pull request, #14407:
URL: https://github.com/apache/cloudstack/pull/14407
### Description
This PR...
<!--- Describe your changes in DETAIL - And how has behaviour functionally changed. -->
<!-- For new features, provide link to FS, dev ML discussion etc. -->
<!-- In case of bug fix, the expected and actual behaviours, steps to reproduce. -->
<!-- When "Fixes: #<id>" is specified, the issue/PR will automatically be closed when this PR gets merged -->
<!-- For addressing multiple issues/PRs, use multiple "Fixes: #<id>" -->
<!-- Fixes: # -->
<!--- ******************************************************************************* -->
<!--- NOTE: AUTOMATION USES THE DESCRIPTIONS TO SET LABELS AND PRODUCE DOCUMENTATION. -->
<!--- PLEASE PUT AN 'X' in only **ONE** box -->
<!--- ******************************************************************************* -->
### Types of changes
- [ ] Breaking change (fix or feature that would cause existing functionality to change)
- [ ] New feature (non-breaking change which adds functionality)
- [ ] Bug fix (non-breaking change which fixes an issue)
- [ ] Enhancement (improves an existing feature and functionality)
- [ ] Cleanup (Code refactoring and cleanup, that may add test cases)
- [ ] Build/CI
- [ ] Test (unit or integration test code)
### Feature/Enhancement Scale or Bug Severity
#### Feature/Enhancement Scale
- [ ] Major
- [ ] Minor
#### Bug Severity
- [ ] BLOCKER
- [ ] Critical
- [ ] Major
- [ ] Minor
- [ ] Trivial
### Screenshots (if appropriate):
### How Has This Been Tested?
<!-- Please describe in detail how you tested your changes. -->
<!-- Include details of your testing environment, and the tests you ran to -->
#### How did you try to break this feature and the system with this change?
<!-- see how your change affects other areas of the code, etc. -->
<!-- Please read the [CONTRIBUTING](https://github.com/apache/cloudstack/blob/main/CONTRIBUTING.md) document -->
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Main fix 13484 14321 [cloudstack]
Posted by "github-actions[bot] (via GitHub)" <gi...@apache.org>.
github-actions[bot] commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6190628647
This pull request has merge conflicts. Dear author, please fix the conflicts and sync your branch with the base branch.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "Damans227 (via GitHub)" <gi...@apache.org>.
Damans227 commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6216091246
> checked again, when a running vm is added to a network with config drive, the ISO will be created , but it is not attached to the running vm, so it has no duplicate id 4 after the NIC is added
@weizhouapache agreed, no duplicate 4. the second iso still moves to 5 on that vm though, and that drive wasnt set up at start, so the attach fails until a restart. ok to add that to the limits in the description?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "weizhouapache (via GitHub)" <gi...@apache.org>.
weizhouapache commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6214865813
### Manual test results (commit de91dd4417a)
**Environment**: management server and one KVM host built from this branch (24.0.0-SNAPSHOT). The agent advertises `host.cdrom.max.count=2`. `vm.iso.max.count` is 1 by default and was raised to 2 for the two-ISO cases (restored afterwards). Template: Ubuntu 24.04 cloud image. Guest OS types: Ubuntu, Windows Server 2022 and "Other (64-bit)".
Every check was asserted by a script (target dev and bus from `virsh dumpxml`, device ids from the DB, VM state after start).
| # | Case | Result |
|---|---|---|
| 1 | BIOS, no config drive, `vm.iso.max.count=1` | Data disks get device ids 1, 2, 5, 6, 7 (`vdb`, `vdc`, `vdf`, `vdg`, `vdh`). A second ISO is rejected. |
| 2 | BIOS, no config drive, 2 ISOs | ISOs on `hdc` and `hdd`. A third ISO is rejected. Data disks 1, 2, 5. |
| 3 | BIOS, config drive, 2 ISOs | ISO1 `hdc`, config drive `hdd`, ISO2 SATA `sdf`. Data disks 1, 2, 6. Detaching ISO2 leaves the config drive intact. |
| 4 | UEFI/q35, SCSI, config drive (#13484) | No duplicate target names. Data disks 1, 2, 6, 7. |
| 5 | UEFI/q35, SCSI, no config drive (#14321) | CD-ROMs `sdd` and `sde`, data disks 1, 2, 5. No clash. |
| 6 | Deployed stopped, 4 disks attached while stopped, config drive | Ids 1, 2, 6, 7. The VM starts cleanly. |
| 7 | Windows Server 2022, secure UEFI | All devices SATA, no clash. ISOs are kept across a restart. Data disks 1, 2, 6, 7, 8. |
| 8 | "Other (64-bit)", BIOS, IDE root | Root `hda`, CD-ROMs `hdc`, `hdd` (config drive) and `sdf`. Data disks 1, 2, 6. |
| 9 | Explicit `deviceid` 3 or 4 on attach | Rejected, the message lists the reserved CD-ROM and config drive ids. |
| 10 | NIC plugged into a running VM that already has config drive | The config drive is re-attached on `hdd`. The CD-ROMs are unchanged. |
| 11 | Import of an unmanaged KVM VM with 4 data disks | Ids 1, 2, 6, 7 (before the change: 1, 2, 3, 4). The VM starts with a valid domain. |
Also checked: a second ISO on a running config drive VM that booted with `vm.iso.max.count=1` fails to attach until the VM is restarted. This matches the description. q35 selected only through `kvm.guest.os.machine.type`, without the `UEFI` detail, still fails to start (IDE is unsupported on q35). This is the same as before the change.
Unit tests of the touched classes pass (`VolumeApiServiceImplTest`, `TemplateManagerImplTest`, `UnmanagedVMsManagerImplTest`, `KVMGuruTest`, `LibvirtComputingResourceTest`, `LibvirtVMDefTest`, `ConfigDriveNetworkElementTest`).
**Not covered by these tests**
- Backup restore (`KVMGuru`), VMware to KVM import, live migration, non-KVM hypervisors, and hosts that advertise a CD-ROM cap other than 2.
- A volume that already sits on a reserved device id is not moved.
- A NIC on a config drive network added to a running VM that booted without config drive: only the ISO file is created, nothing is attached until the next start. ISO attach/detach on that VM before the restart was not tested.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "andrijapanicsb (via GitHub)" <gi...@apache.org>.
andrijapanicsb commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6352081598
Thanks, @weizhouapache , for working through these conflicts, and @Damans227 for highlighting the remaining runtime limitation.
Could we also revisit whether the second user ISO drive introduced in #13101 is worth retaining? My concern is the maintenance and regression cost relative to the requirements it addresses.
For an attended Windows installation, a second CD-ROM is not required: the existing single-drive workflow is to swap the Windows ISO for the VirtIO ISO, load the storage driver, then restore the Windows ISO before continuing (like it was done for last 30 years or so, on physical HW). The original PR itself acknowledges that workflow. Two drives make this more convenient, but I would question whether that convenience alone justifies the additional complexity now needed around device allocation, config-drive coexistence, controller selection, and runtime attachment.
One particularly concerning detail is that the KVM agent pre-allocates the extra CD-ROM even with `vm.iso.max.count=1`. As #14321 demonstrates, an empty drive can then conflict with a data disk. That extends the impact to users who have not opted into attaching multiple ISOs.
I recognise the convenience that #13101 brings, but I would welcome concrete requirements beyond avoiding the manual swap before making a decision.
My preference would be to reconsider the feature and, unless those requirements justify the ongoing cost, plan its deprecation and removal while retaining one user ISO drive alongside the separate config drive. Since it has shipped in 4.23, that would need a compatibility plan for existing attachments and API consumers. If we retain it, I think the minimum should be genuinely optional extra drives and predictable device assignments across config-drive changes and restarts.
The older config-drive/data-volume conflict in #13484 still needs fixing regardless. I appreciate the work in this PR; I would just like us to assess whether supporting the additional user ISO drive remains the right trade-off.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "weizhouapache (via GitHub)" <gi...@apache.org>.
weizhouapache commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6199783136
> overall the changes look good.
>
> nit: what happens if a running vm is added to a config drive network and then gets a second iso? looks like the iso goes to a drive that was never set up at start, so the attach fails until a restart.
good point.
if so, I would advise to reserve device id 3 as primary CD-ROM, and device id 4 as configdrive ISO.
if the VM use IDE controller which support 2 CDROMs, the additional CD-ROMs (from 2nd CDROM) will use SATA controller, due to hardcoded `MAX_CDROMS_PER_VM = 2` , right ?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "blueorangutan (via GitHub)" <gi...@apache.org>.
blueorangutan commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6451791720
@RosiKyu a [SL] Jenkins job has been kicked to build packages. It will be bundled with no SystemVM templates. I'll keep you posted as I make progress.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "weizhouapache (via GitHub)" <gi...@apache.org>.
weizhouapache commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6412940763
> Thanks, @weizhouapache , for working through these conflicts, and @Damans227 for highlighting the remaining runtime limitation.
>
> Could we also revisit whether the second user ISO drive introduced in #13101 is worth retaining? My concern is the maintenance and regression cost relative to the requirements it addresses.
>
> For an attended Windows installation, a second CD-ROM is not required: the existing single-drive workflow is to swap the Windows ISO for the VirtIO ISO, load the storage driver, then restore the Windows ISO before continuing (like it was done for last 30 years or so, on physical HW). The original PR itself acknowledges that workflow. Two drives make this more convenient, but I would question whether that convenience alone justifies the additional complexity now needed around device allocation, config-drive coexistence, controller selection, and runtime attachment.
>
> One particularly concerning detail is that the KVM agent pre-allocates the extra CD-ROM even with `vm.iso.max.count=1`. As #14321 demonstrates, an empty drive can then conflict with a data disk. That extends the impact to users who have not opted into attaching multiple ISOs.
>
> I recognise the convenience that #13101 brings, but I would welcome concrete requirements beyond avoiding the manual swap before making a decision.
>
> My preference would be to reconsider the feature and, unless those requirements justify the ongoing cost, plan its deprecation and removal while retaining one user ISO drive alongside the separate config drive. Since it has shipped in 4.23, that would need a compatibility plan for existing attachments and API consumers. If we retain it, I think the minimum should be genuinely optional extra drives and predictable device assignments across config-drive changes and restarts.
>
> The older config-drive/data-volume conflict in #13484 still needs fixing regardless. I appreciate the work in this PR; I would just like us to assess whether supporting the additional user ISO drive remains the right trade-off.
cc @Damans227 @NuxRo
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "ACSHomeBot (via GitHub)" <gi...@apache.org>.
ACSHomeBot commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6427491339
### Packaging results:
| Result | Artifact | Platform | Build Time |
|:--|:--|:--|--:|
| PASS | RPM | EL (EL8/9/10) | 34 min |
| PASS | DEB | Ubuntu, Debian | 22 min |
**Test packages are available at:**
- EL9/RPM: [https://packages.v2kmigrate.com/cloudstack/pr/14407/el9/](https://packages.v2kmigrate.com/cloudstack/pr/14407/el9/index.html)
- Debian/APT: [https://packages.v2kmigrate.com/cloudstack/pr/14407/debian/](https://packages.v2kmigrate.com/cloudstack/pr/14407/debian/index.html)
<!-- acshomebot-package-results:20261026T201124Z-f9451683:1 -->
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "vishesh92 (via GitHub)" <gi...@apache.org>.
vishesh92 commented on code in PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#discussion_r4282992746
##########
server/src/main/java/com/cloud/template/TemplateManagerImpl.java:
##########
@@ -714,6 +726,60 @@ public void prepareIsoForVmProfile(VirtualMachineProfile profile, DeployDestinat
}
}
+ @Override
Review Comment:
The changes in this file looks good but I am not sure if this is the place where these methods should exist.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "github-actions[bot] (via GitHub)" <gi...@apache.org>.
github-actions[bot] commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6400121099
This pull request has merge conflicts. Dear author, please fix the conflicts and sync your branch with the base branch.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "Damans227 (via GitHub)" <gi...@apache.org>.
Damans227 commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6197617368
overall the changes look good.
nit: what happens if a running vm is added to a config drive network and then gets a second iso? looks like the iso goes to a drive that was never set up at start, so the attach fails until a restart.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "RosiKyu (via GitHub)" <gi...@apache.org>.
RosiKyu commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6451759440
@blueorangutan package
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "Damans227 (via GitHub)" <gi...@apache.org>.
Damans227 commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6201655948
> if the VM use IDE controller which support 2 CDROMs, the additional CD-ROMs (from 2nd CDROM) will use SATA controller or 2nd IDE controller, due to hardcoded `MAX_CDROMS_PER_VM = 2` , right ?
@weizhouapache right, the limit of 2 is ours, not qemu's. with 4 always kept for the config drive, the ide bus is full after the first cd, so extra isos go on sata or a second ide controller. that also covers the running vm case, since the second iso no longer moves when a config drive shows up.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "blueorangutan (via GitHub)" <gi...@apache.org>.
blueorangutan commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6414271353
Packaging result [SF]: ✖️ el8 ✖️ el9 ✖️ debian ✖️ suse15. SL-JID 19743
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "ACSHomeBot (via GitHub)" <gi...@apache.org>.
ACSHomeBot commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6426219060
Accepted: `package kvm G` for `3336b483e8e7`.
Planned: **RPM + DEB**, **KVM SystemVM**.
<!-- acshomebot-package-state:20261026T201124Z-f9451683 -->
Status: **Queued** for **worker G**. Jobs ahead: **0**.
<!-- /acshomebot-package-state:20261026T201124Z-f9451683 -->
<!-- acshomebot-package-status:20261026T201124Z-f9451683 -->
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "andrijapanicsb (via GitHub)" <gi...@apache.org>.
andrijapanicsb commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6426215161
@acshomebot package G
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "codecov[bot] (via GitHub)" <gi...@apache.org>.
codecov[bot] commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6190950293
## [Codecov](https://app.codecov.io/gh/apache/cloudstack/pull/14407?dropdown=coverage&src=pr&el=h1&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache) Report
:white_check_mark: All modified and coverable lines are covered by tests.
:white_check_mark: Project coverage is 3.70%. Comparing base ([`d3f0700`](https://app.codecov.io/gh/apache/cloudstack/commit/d3f0700e978390f65805a488a96ef9ca8e9e8070?dropdown=coverage&el=desc&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache)) to head ([`3076489`](https://app.codecov.io/gh/apache/cloudstack/commit/30764893a9d72d5fe1f1526585bbf1c9fc588088?dropdown=coverage&el=desc&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache)).
> :exclamation: There is a different number of reports uploaded between BASE (d3f0700) and HEAD (3076489). Click for more details.
>
> <details><summary>HEAD has 1 upload less than BASE</summary>
>
>| Flag | BASE (d3f0700) | HEAD (3076489) |
>|------|------|------|
>|unittests|1|0|
></details>
<details><summary>Additional details and impacted files</summary>
```diff
@@ Coverage Diff @@
## main #14407 +/- ##
=============================================
- Coverage 20.39% 3.70% -16.70%
=============================================
Files 6481 493 -5988
Lines 588040 42544 -545496
Branches 72021 8065 -63956
=============================================
- Hits 119957 1577 -118380
+ Misses 455052 40741 -414311
+ Partials 13031 226 -12805
```
| [Flag](https://app.codecov.io/gh/apache/cloudstack/pull/14407/flags?src=pr&el=flags&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache) | Coverage Δ | |
|---|---|---|
| [uitests](https://app.codecov.io/gh/apache/cloudstack/pull/14407/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache) | `3.70% <ø> (ø)` | |
| [unittests](https://app.codecov.io/gh/apache/cloudstack/pull/14407/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache) | `?` | |
Flags with carried forward coverage won't be shown. [Click here](https://docs.codecov.io/docs/carryforward-flags?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache#carryforward-flags-in-the-pull-request-comment) to find out more.
</details>
[:umbrella: View full report in Codecov by Harness](https://app.codecov.io/gh/apache/cloudstack/pull/14407?dropdown=coverage&src=pr&el=continue&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache).
:loudspeaker: Have feedback on the report? [Share it here](https://about.codecov.io/codecov-pr-comment-feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache).
<details><summary> :rocket: New features to boost your workflow: </summary>
- :snowflake: [Test Analytics](https://docs.codecov.com/docs/test-analytics): Detect flaky tests, report on failures, and find test suite problems.
- :package: [JS Bundle Analysis](https://docs.codecov.com/docs/javascript-bundle-analysis): Save yourself from yourself by tracking and limiting bundle sizes in JS merges.
</details>
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "weizhouapache (via GitHub)" <gi...@apache.org>.
weizhouapache commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6210613792
> > nit: what happens if a running vm is added to a config drive network and then gets a second iso? looks like the iso goes to a drive that was never set up at start, so the attach fails until a restart.
checked again, when a running vm is added to a network with config drive, the ISO will be created , but it is not attached to the running vm, so it has no duplicate id 4 after the NIC is added
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "RosiKyu (via GitHub)" <gi...@apache.org>.
RosiKyu commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6464896774
Tested on build 4.22.2.0-shapeblue22281 (head 7afd3e96951), upgraded from 4.22.1.0. The q35 hosts use config drive and NAS backup. The baseline was also checked on 4.22.1.1.
**LGTM for the reported issue:** on q35 config drive VMs, the 3rd data volume no longer takes deviceid 4, on running and stopped VMs. No regression was found.
### Tested/Verified
- The bug reproduces on 4.22.1.0, on both running and stopped VMs.
- PR build, q35 scsi and virtio config drive VMs: data volumes get 1, 2, 5, and the VMs start.
- VMs without config drive: no change, deviceid 4 is still used.
- Explicit device ids: 3 and 4 are rejected with the new message. Out of range ids are rejected, and free ids are accepted.
- Limits and lifecycle:
- the 32-volume limit is still reachable;
- after a detach, the lowest free id is reused;
- stop/start, reboot and live migration keep the device ids;
- with an ISO attached, the next volume skips 3 and 4;
- after a config drive NIC is added, the next volume gets 5.
- Upgrade:
- a new management server with old agents already gives 5;
- existing i440fx VMs with a volume on 4 keep working.
- Behaviour change: an explicit deviceid=4 on i440fx config drive VMs is now rejected.
- NAS backup: create a VM from a backup and restore in place both keep the ids, and the VM starts.
- Unmanaged import into a config drive network: data volumes get 1, 2, 5, and the VM starts.
### Potential issues identified
1. A VM created from a backup reuses the backup's deviceid 4 on a config drive network. Take a backup of a VM without config drive that has its 3rd data volume on deviceid 4. Then create a VM from that backup on a config drive network. The volume keeps deviceid 4, and the VM can't start (target 'sde' duplicated). It reproduced 4 of 4 times. It also fails on 4.22.1.1, so the PR didn't cause it. The same backup on a network without config drive gives a running VM. Creating the VM with explicit device ids 1, 2, 5 also gives a running VM.
2. Known limitation from the Notes, confirmed: a config drive VM that already has a data volume on deviceid 4 still can't start. This covers VMs upgraded from 4.22.1.0 and VMs that get a config drive NIC later. After a detach and attach, the volume gets deviceid 5 and the VM starts.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "blueorangutan (via GitHub)" <gi...@apache.org>.
blueorangutan commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6412981455
@weizhouapache a [SL] Jenkins job has been kicked to build packages. It will be bundled with no SystemVM templates. I'll keep you posted as I make progress.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "weizhouapache (via GitHub)" <gi...@apache.org>.
weizhouapache commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6412947467
@blueorangutan package
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "RosiKyu (via GitHub)" <gi...@apache.org>.
RosiKyu merged PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "RosiKyu (via GitHub)" <gi...@apache.org>.
RosiKyu commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6464958886
Tested on build 24.0.0-shapeblue22317 (head 3336b483e8e), upgraded from 4.23.0.0: first the management server, then the agents. The q35 hosts (host.cdrom.max.count 2) use config drive and NAS backup.
LGTM for the reported issues (#13484 and #14321): data volumes no longer take the device ids of the CD-ROMs or the config drive ISO, on running and stopped VMs, with 1 and 2 ISOs. No regression was found.
### Tested/Verified
- Bugs reproduced on 4.23.0.0:
- the 3rd data volume collides on sde with the config drive ISO, or with the empty 2nd CD-ROM;
- on stopped VMs it gets deviceid 4 and the VM can't start;
- with vm.iso.max.count=2, no new VM on a config drive network could start.
- PR build, q35 scsi and virtio VMs: data volumes get 1, 2, 6, 7 with config drive and 1, 2, 5, 6 without. The VMs start.
- Explicit device ids: reserved ids are rejected with the new message ([3, 4, 5] with config drive, [3, 4] without). Out of range ids are rejected, and free ids are accepted.
- 32-volume limit: still reachable (1, 2, 6 to 35).
- vm.iso.max.count=2, i440fx and q35: ISO 1, the config drive and ISO 2 are on 3, 4 and 5 (ISO 2 on sata sdf). Detaching ISO 2 leaves the config drive, and stop/start keeps all 3 CD-ROMs.
- Lifecycle: reboot and live migration keep the device ids and targets.
- Notes confirmed:
- VMs with a volume already on a reserved id can't start until the volume is detached and attached;
- ISO 2 on a running config drive VM needs a restart if the VM was started with vm.iso.max.count=1.
- Upgrade:
- a new management server with 4.23 agents already gives 1, 2, 6;
- existing i440fx VMs with volumes on 4 and 5 keep working;
- an ISO in the 2nd slot moves to sdf once the VM gets a config drive NIC.
- Behaviour change: an explicit deviceid=5 on config drive VMs was accepted on 4.23.0.0 and is now rejected.
- Other paths:
- deploy with a data disk offering;
- create a VM from a backup of a config drive VM;
- restore in place;
- unmanaged import with 4 data disks (1, 2, 6, 7).
### Potential issues (both only with vm.iso.max.count=2)
1. A VM without config drive that later gets a config drive NIC keeps its data volume on deviceid 5. The PR build itself assigned that id. At setting 1 the VM starts. At 2 it can't start, because the volume and the ISO 2 CD-ROM both get sdf.
2. A VM created from a backup keeps the backup's deviceid 5 on a config drive network. At setting 1 the VM starts; at 2 it can't start (target 'sdf' duplicated). The same backup on a network without config drive gives a running VM. This is the same path as the open backup case on #14073. It was not checked on 4.23.0.0.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "blueorangutan (via GitHub)" <gi...@apache.org>.
blueorangutan commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6453406933
Packaging result [SF]: ✔️ el8 ✔️ el9 ✔️ el10 ✔️ debian ✔️ suse15. SL-JID 19780
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] server,kvm: Avoid device id conflicts between data volumes, CD-ROMs and the config drive ISO [cloudstack]
Posted by "RosiKyu (via GitHub)" <gi...@apache.org>.
RosiKyu commented on PR #14407:
URL: https://github.com/apache/cloudstack/pull/14407#issuecomment-6468257761
Follow-up issues for the open cases from testing. All three also fail on 4.23.0.0 (before this PR), so they are not caused by it:
- #14512 VM created from a backup keeps the backup's data volume device ids on a config drive network
- #14513 VM that gets a NIC on a config drive network keeps a data volume on a reserved device id
- #14514 VMs with a data volume already on a reserved device id still cannot start after the upgrade
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@cloudstack.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org