Sync feature limit vnc console with master #6819
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
To avoid CI format check error
Two commands are used to set max_cstate: xenpm to set at runtime and xen-cmdline to set it in grub conf file to take effect after reboot. Signed-off-by: Changlei Li <[email protected]>
String is used to represent the max_cstate and max_sub_cstate. "" -> unlimited "N" -> max cstate CN "N,M" -> max cstate CN with max sub state M Just follow the xen-cmdline cstate, see https://xenbits.xen.org/docs/unstable/misc/xen-command-line.html#max_cstate-x86 Signed-off-by: Changlei Li <[email protected]>
Signed-off-by: Changlei Li <[email protected]>
C-states are power management states for CPUs where higher numbered
states represent deeper sleep modes with lower power consumption but
higher wake-up latency. The max_cstate parameter controls the deepest
C-state that CPUs are allowed to enter.
Common C-state values:
- C0: CPU is active (not a sleep state)
- C1: CPU is halted but can wake up almost instantly
- C2: CPU caches are flushed, slightly longer wake-up time
- C3+: Deeper sleep states with progressively longer wake-up times
To set max_cstate on dom0 host, two commands are used: `xenpm` to set at
runtime and `xen-cmdline` to set it in grub conf file to take effect
after reboot.
xenpm examples:
```
# xenpm set-max-cstate 0 0
max C-state set to C0
max C-substate set to 0 succeeded
# xenpm set-max-cstate 0
max C-state set to C0
max C-substate set to unlimited succeeded
# xenpm set-max-cstate unlimited
max C-state set to unlimited
# xenpm set-max-cstate -1
Missing, excess, or invalid argument(s)
```
xen-command-line examples:
```
/opt/xensource/libexec/xen-cmdline --get-xen max_cstate
"" -> unlimited
"max_cstate=N" -> max cstate N
"max_cstate=N,M" -> max cstate N, max c-sub-state M *)
/opt/xensource/libexec/xen-cmdline --set-xen max_cstate=1
/opt/xensource/libexec/xen-cmdline --set-xen max_cstate=1,0
/opt/xensource/libexec/xen-cmdline --delete-xen max_cstate
```
[xen-command-line.max_cstate](https://xenbits.xen.org/docs/unstable/misc/xen-command-line.html#max_cstate-x86).
This PR adds a new field `host.max_cstate` to manage host's max_cstate.
`host.set_max_cstate` use the two commands mentioned above to configure.
While dbsync on xapi start, the filed will be synced by `xen-cmdline
--get-xen max_cstate`
Signed-off-by: Changlei Li <[email protected]>
- write ntp servers to chrony.conf - interaction with dhclient - handle /run/chrony-dhcp/$interface.sources - handle chrony.sh - restart/enable/disable chronyd Signed-off-by: Changlei Li <[email protected]>
Signed-off-by: Changlei Li <[email protected]>
Signed-off-by: Changlei Li <[email protected]>
At XAPI start, check the actual NTP config to determine the ntp mode, ntp enabled, ntp custom servers and store in xapi DB. Signed-off-by: Changlei Li <[email protected]>
Signed-off-by: Changlei Li <[email protected]>
Signed-off-by: Changlei Li <[email protected]>
New filed: `host.ntp_mode`, `host.ntp_custom_servers` New API: `host.set_ntp_mode`, `host.set_ntp_custom_servers`, `host.get_ntp_mode`, `host.get_ntp_custom_servers`, `host.get_ntp_servers_status`. **ntp_mode_dhcp**: In this mode, ntp uses the dhcp assigned ntp servers as sources. In Dom0, dhclient triggers `chrony.sh` to update the ntp servers when network event happens. It writes ntp servers to `/run/chrony-dhcp/$interface.sources` and the dir `/run/chrony-dhcp` is included in `chrony.conf`. The dhclient also stores dhcp lease in `/var/lib/xcp/dhclient-$interface.leases`, see https://github.com/xapi-project/xen-api/blob/v25.31.0/ocaml/networkd/lib/network_utils.ml#L925. When switch ntp mode to dhcp, XAPI checks the lease file and finds ntp server then fills chrony-dhcp file. The exec permission of `chrony.sh` is added. When swith ntp mode from dhcp to others, XAPI removes the chrony-dhcp files and the exec permission of `chrony.sh`. The operation is same with xsconsole https://github.com/xapi-project/xsconsole/blob/v11.1.1/XSConsoleData.py#L593. In this feature, xsconsole will change to use XenAPI to manage ntp later to avoid conflict. **ntp_mode_custom**: In this mode, ntp uses `host.ntp_custom_servers` as sources. This is implemented by changing `chrony.conf` and restart chronyd. `host.ntp_custom_servers` is set by the user. **ntp_mode_default**: In this mode, ntp uses default-ntp-servers in XAPI config file.
For example, the legacy default ntp servers are [0-3].centos.pool.ntp.org, and current default ntp servers are [0-3].xenserver.pool.ntp.org. After update or upgrade, the legacy default ntp servers are recognized and changed to current default ntp servers. The mode is ntp_mode_default as well. Signed-off-by: Changlei Li <[email protected]>
For example, the legacy default ntp servers are `[0-3].centos.pool.ntp.org`, and current default ntp servers are `[0-3].xenserver.pool.ntp.org`. After update or upgrade, the legacy default ntp servers are recognized and changed to current default ntp servers. The mode is `ntp_mode_default` as well. Add a new config option named legacy-default-ntp-servers. It will be defined in xapi.conf.d/xenserver.conf (the same with default-ntp-servers)
Hugo supports three styles of front matter to add meta data. See https://gohugo.io/content-management/front-matter/. Follow other pages in content/design/ to use yaml style. Signed-off-by: Changlei Li <[email protected]>
Hugo supports three styles of front matter to add meta data. See https://gohugo.io/content-management/front-matter/. Follow other pages in content/design/ to use yaml style.
Signed-off-by: Changlei Li <[email protected]>
In the case of cross pool SXM, the running VM will be created on destination host with initial halted state. The state shall be changed by refresh_vm. The function force_state_reset before it will incorrectly clear some vm config. Move force_state_reset after refresh_vm fater refresh_vm in pool_migrate_complete to solve it. Signed-off-by: Changlei Li <[email protected]>
…api-project#6731) There is a regression test fail between xapi v25.30.0 and v25.33.0. The job is cross pool SXM with vGPU. The source host VM.migrate_send failed with exception `Storage_error ([S(Does_not_exist);[S(mirror)` which is raised by `MIRROR.stat`. The `MIRROR.stat` is triggered by destination host in `pool_migrate_complete`. The error is `Server_error(HANDLE_INVALID, [ PGPU; OpaqueRef:NULL ])`. I find xapi-project#6648 inserts `force_state_reset_keep_current_operations` in `pool_migrate_complete` which set VGPU resident_on to NULL. After reverting this commit, the job pass.
Although I doubt it, please check whether Xenserver's proprietary code uses these functions. Signed-off-by: Pau Ruiz Safont <[email protected]>
This makes changing the chop tests easier Signed-off-by: Pau Ruiz Safont <[email protected]>
This function is equivalent to calling take and drop on the same parameters; as well as being very similar to the chop function, with the difference that when using an out-of-bounds parameter now it returns empty lists instead of raising an exception. Signed-off-by: Pau Ruiz Safont <[email protected]>
Now users can use out-of-bounds limits without fear. I've used the opportunity to consolidate code that sorts and splits lists in db_gc_util. Signed-off-by: Pau Ruiz Safont <[email protected]>
This can be replaced with a List.drop. For now this is available in xapi's stdext, but it will become part of the standard library Signed-off-by: Pau Ruiz Safont <[email protected]>
It's surprising this bug went unnoticed Signed-off-by: Pau Ruiz Safont <[email protected]>
Signed-off-by: Guillaume <[email protected]>
There is race condition about vm cache between pool_migrate_complete and VM event. In the cross-pool migration case, it is designed to create vm with power_state Halted in XAPI db. In pool_migrate_complete, add_caches create an empty xenops_chae for the VM, then refresh_vm compares the cache powerstate None with its real state Running to update the right powerstate to XAPI db. In the fail case, it is found that: -> VM event 1 update_vm -> pool_migrate_complete add_caches (cache power_state None) -> pool_migrate_complete refresh_vm -> VM event 1 update cache (cache power_state Running) -> VM event 2 update_vm (Running <-> Running, XAPI DB not update) When pool_migrate_complete add_caches, the cache update of previous VM event 1 breaks the design intention. This commit add a wait in pool_migrate_complete to ensure all in-flight events complete before add_caches. Then there will be no race condition. Signed-off-by: Changlei Li <[email protected]>
There is race condition about vm cache between pool_migrate_complete and VM event. In the cross-pool migration case, it is designed to create vm with power_state Halted in XAPI db. In pool_migrate_complete, add_caches create an empty xenops_chae for the VM, then refresh_vm compares the cache powerstate None with its real state Running to update the right powerstate to XAPI db. In the fail case, it is found that: -> VM event 1 update_vm -> pool_migrate_complete add_caches (cache power_state None) -> pool_migrate_complete refresh_vm -> VM event 1 update cache (cache power_state Running) -> VM event 2 update_vm (Running <-> Running, XAPI DB not update) When pool_migrate_complete add_caches, the cache update of previous VM event 1 breaks the design intention. This commit add a wait in pool_migrate_complete to ensure all in-flight events complete before add_caches. Then there will be no race condition.
This function is useful to apply a function to a list that may fail, it returns the first error, or all the successful results. Signed-off-by: Pau Ruiz Safont <[email protected]>
Now Db_not_initialized is raised instead of Not_found, which was raised by List.hd. This makes the exception much more recognizable and easier to locate. Signed-off-by: Pau Ruiz Safont <[email protected]>
These can be easily replaced with a match against the list, or changing parameters to separate the head from the rest of the list Signed-off-by: Pau Ruiz Safont <[email protected]>
Previously an exception was raised for last, head is a new function. Also adds a latest_release to datamodel_types that can't fail after the module has been loaded. (the list is populated so it won't fail when the module is loaded either) Signed-off-by: Pau Ruiz Safont <[email protected]>
This forces users to deal with None in the new code. One of them is replace with the new function List.try_map that returns the first error. While split_on_char never returns an empty list, it's a good test for using try_map Signed-off-by: Pau Ruiz Safont <[email protected]>
The return type is the same, but it's more obvious to see what it does Signed-off-by: Pau Ruiz Safont <[email protected]>
This is done by using options and avoiding exceptions as much as possible, or using custom exceptions that help finding the cause.
Currently, vhd-tool provides several "hybrid" modes where it exports into vhd from raw, using the information from the VHD bitmaps to determine which blocks and sectors contain data (to avoid reading zero blocks). Other tools are also handling VHD-backed VDIs (we are exporting them as part of XVA export, and now they can also be exported to QCOW), and currently they have to read the whole raw disk. Instead provide a read_headers command which provides data on allocated clusters for other tools to use, allowing them to speed up handling sparse VDIs. It uses a new blocks_json function in Vhd_format. Signed-off-by: Andrii Sultanov <[email protected]>
The body has less indentation this way Signed-off-by: Andrii Sultanov <[email protected]>
This allows using it in stream_vdi and qcow_tool_wrapper without introducing a dependency cycle. Signed-off-by: Andrii Sultanov <[email protected]>
Qcow_tool_wrapper and Vhd_tool_wrapper expect a particular driver to be backing the VDI and fall back to handling the VDI as raw otherwise - they will be using backing_file_of_device_with_driver. Stream_vdi, however, will need to branch on the type of the driver, and it will use backing_info_of_device (which also returns the type of the driver) Signed-off-by: Andrii Sultanov <[email protected]>
Split common code used by {Vhd,Qcow}_tool_wrapper into a new vhd_qcow_parsing
module.
Since Vhd_tool_wrapper.run_vhd_tool is hardcoded to read the progress
percentage printed by vhd-tool, we have to use the more generic
Vhd_qcow_parsing.run_qcow_tool to run vhd-tool.
Since VHD and QCOW follow the same format of JSON, use the same parse_header
function.
Signed-off-by: Andrii Sultanov <[email protected]>
Reads the bitmaps for VHD- and QCOW-backed VDIs, determines which clusters are allocated and only reads and writes these to the resulting xva. This avoids the need for the "timeout workaround", which is needed when no data has been sent for an extended period of time (so stream_vdi writes a "packet" that doesn't carry any data, just a checksum of an empty body. in case of a compressed export, however, the compressor binary buffers output and this timeout workaround does not work). This also greatly speeds up export of VMs with sparse VDIs. Signed-off-by: Andrii Sultanov <[email protected]>
Some of the users of the function did not handle exceptions correctly - make the "not found" case explicit with an option. Signed-off-by: Andrii Sultanov <[email protected]>
Signed-off-by: Andrii Sultanov <[email protected]>
Signed-off-by: Andrii Sultanov <[email protected]>
…xport (xapi-project#6786) Following xapi-project#6769, add a `read_headers` command to `vhd-tool` (following `qcow-tool`'s JSON format). This allows `stream_vdi` to determine which clusters are allocated in QCOW- and VHD-backed VDIs, only reading and writing allocated blocks (previously it read the whole raw disk, verifying if blocks only contain zeros). If there are any issues during header parsing, it falls back to the slow path (we don't handle errors during XVA export well, embedding 500 packets inside 200s) This greatly speeds up XVA export for VMs with sparse VDIs: 5gb empty VDI: 19s -> 3s 5gb empty VDI + 2mb filled VDI: 22s -> 6s 5gb empty VDI + half-empty VDI (~4 gigs out of 10): 89s -> 49s Note: If the block size of the VDI is larger than the size of the XVA blocks (this is currently the case for VHD, it has blocks of 2mb, while stream_vdi splits xvas into files of 1mb), stream_vdi can overestimate the size of allocated data (say, if only the first half of the VHD block has data), but in testing with real VDIs this impact was within a margin of error.
Signed-off-by: Stephen Cheng <[email protected]>
stephenchengCloud
requested review from
BengangY,
changlei-li,
gangj and
minglumlu
Contributor
|
Remember to bump schema_minor_vsn as the comment says |
changlei-li
approved these changes
Jan 7, 2026
BengangY
approved these changes
Jan 7, 2026
Signed-off-by: Stephen Cheng <[email protected]>
Collaborator
Author
Updated vsn and also the datamodel lifecycle |
stephenchengCloud
merged commit 64fda7a
into
xapi-project:feature/limit-vnc-console-sessions
16 checks passed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
18 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Sync feature branch with master. No code changed.