fanhuadesenlinnn/cloudpods
1
0
Fork 0
mirror of https://hubproxy.babadafafafafa.cn/https://github.com/yunionio/cloudpods.git synced 2026-09-20 08:03:53 +08:00
Code Issues Packages Projects Releases Wiki Activity
20,836 Commits 27 Branches 687 Tags
d39bf748519c631a356f63be8ec64a74483465f7
Commit Graph

20836 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
wanyaoqi
d39bf74851 fix(host): guest desc correct rbd disk path (#25394) 2026-08-24 14:36:57 +08:00
wanyaoqi
dff7fdf3ed fix(region,host): host register with correct hostname (#25206) 
* fix(region,host): host register with correct hostname

* fix(region): guest migrate reuse create params schedtag
 2026-07-28 10:59:19 +08:00
Jian Qiu
06c8d58ef0 fix: allow empty ip on link-local nic (#25199) 
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
 2026-07-22 11:05:07 +08:00
Jian Qiu
ee0522003c fix: lbagent service code typo 20260604 (#24975) 
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
v4.0.3-20260606.0 		2026-06-05 01:09:46 +08:00
Jian Qiu
fc361cd558 fix: compatibility issue with vms of empty hypervisor (#24971) 
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
v4.0.3-20260604.0 		2026-06-04 15:05:52 +08:00
屈轩
0ae741ddd9 fix(region): purge external project (#24938) v4.0.3-20260602.0 2026-05-31 14:22:48 +08:00
屈轩
cd22014bd4 fix(region): init cloudpods host type (#24937) 2026-05-31 14:22:32 +08:00
Zexi Li
ba3601b846 feat(monitor): add ResourceMetricsDefaultInterval option (#24907) v4.0.3-20260527.0 2026-05-27 11:50:35 +08:00
Jian Qiu
bde5de9ce8 fix: throttle image cache progress callback to avoid download bottleneck (#24899) 
When using S3/MinIO backend, the progress callback in AcquireImage was
called synchronously on every 4KB read chunk (via StreamPipe2). Each
call invokes UpdateServerProgress which makes a blocking HTTP PUT to the
compute API (~50ms), capping download speed at ~0.65 Mbps regardless of
available bandwidth.

Fix: throttle the callback to fire at most once every 5 seconds and make
the HTTP call asynchronous (goroutine), matching the behavior of v3.11.x
which used a 1-second ticker with async callbacks.

Fixes #24897

Co-authored-by: Mrliuch <liu15094534492@gmail.com>
 2026-05-26 15:09:29 +08:00
屈轩
af9136e080 fix(region): skip useless addition network wire create (#24895) v4.0.3-20260526.0 2026-05-25 19:46:04 +08:00
Zexi Li
acee027365 fix(host): aggregate same-type container devices for env config (#24890) 
When multiple NVIDIA GPUs were attached to a container, each device was
passed individually to GetContainerExtraConfigures, producing one
NVIDIA_VISIBLE_DEVICES env per GPU. Duplicate env keys are collapsed by
the runtime to the last value, so the container only saw one GPU even
though the spec listed all of them.

Group RestDevs by DeviceType and invoke GetContainerExtraConfigures
once per type so all GPU UUIDs are joined into a single
NVIDIA_VISIBLE_DEVICES value.
 2026-05-25 17:29:42 +08:00
Jian Qiu
cc346ba983 fix: disable s3 direct download default (#24888) 
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
v4.0.3-20260525.0 		2026-05-25 10:04:21 +08:00
Zexi Li
4e41809569 fix(region): prefer check_code series for lb backend check_status (#24879) 2026-05-22 14:19:40 +08:00
Jian Qiu
197c91a014 fix: set bridge up after created (#24872) 
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
v4.0.3-20260521.1 		2026-05-21 20:28:35 +08:00
wanyaoqi
37f6cd57a4 fix(scheduler): set options ContainerNumaAllocate default true (#24871) 2026-05-21 19:32:23 +08:00
Zexi Li
819f22dd99 feat(hermes): connect vllm/sglang/ollama (#24864) 
Co-authored-by: cwz <cwz_eikoh@163.com>
v4.0.3-20260521.0 		2026-05-21 11:41:39 +08:00
Zexi Li
4432749b7b feat(llm): Implement resumable downloads for models from Hugging Face (#24860) 
Co-authored-by: cwz <cwz_eikoh@163.com>
 2026-05-21 10:48:56 +08:00
Zexi Li
588e282571 fix(region): ensure lb backend check_status returned when check_code missing (#24857) 2026-05-21 10:34:17 +08:00
wanyaoqi
42b85ba2c7 fix(host): remove cgroup root path check quotation marks (#24853) 2026-05-21 10:33:45 +08:00
Zexi Li
4401be272c fix(region): query metrics from monitor service (#24839) v4.0.3-20260519.0 2026-05-19 18:02:27 +08:00
Zexi Li
827c6b90dd fix(llm): persistent /config dir of hermes agent (#24836) 2026-05-18 14:57:59 +08:00
Zexi Li
3c90e78377 feat(climc): add generic feature-config command (#24831) 
Add a generic `feature-config <feature> <on|off>` command alongside the
existing per-feature `feature-config-<name>` commands. A `--force` flag
skips the known-features validation so new features can be toggled
without a code change. Also adds "ai" to the known feature list.
 2026-05-18 14:29:04 +08:00
Zexi Li
fd1767cebe feat(llm): support sglang (#24823) 
Co-authored-by: cwz <cwz_eikoh@163.com>
v4.0.3-20260514.0 v4.0.3-20260515.0 v4.0.3-20260518.0 v4.0.3-20260518.1 		2026-05-14 17:08:53 +08:00
Jian Qiu
2d8e88aae7 fix: lbagent keepalived service run may exit (#24806) 
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
 2026-05-13 12:07:54 +08:00
屈轩
d7dc96ac00 fix(region): support stop for change vm config (#24817) 2026-05-12 19:48:09 +08:00
屈轩
d1bbf45990 fix(region): update gp3 disk size (#24812) 2026-05-12 17:27:34 +08:00
Zexi Li
79ab73c980 Automated cherry pick of #24801: Feature(vllm): use entrypint instead of StartLLM (#24802) 
* feat(llm): use entrypoint instead of StartLLM

* feat(llm): adapt front-end in comfyui instant-model

---------

Co-authored-by: cwz <cwz_eikoh@163.com>
 2026-05-12 10:46:21 +08:00
Zexi Li
680dd5a2dc fix(llm): honor SLLM device override when building container spec (#24800) 2026-05-11 20:06:58 +08:00
wanyaoqi
f63e35af99 fix(host): remove -no-kvm args, -no-kvm argument was a synonym for setting -machine accel=tcg (#24789) 2026-05-11 20:01:14 +08:00
wanyaoqi
505ef9e5bb fix(host): check is cpu hotplug support on guest init (#24787) 2026-05-11 19:59:14 +08:00
Zexi Li
c1abc7570d feat(llm): add SLLM mounted_models override and driver-level validation (#24796) 2026-05-11 17:41:16 +08:00
Zexi Li
c28edb026e feat(monitor): use GET for unifiedmonitors resource-metrics (#24782) 2026-05-08 16:19:35 +08:00
Zexi Li
b5cf4bb575 fix(monitor): dedupe alert trigger condition in notify templates (#24780) 2026-05-07 15:52:51 +08:00
屈轩
c35e95046d fix(region): avoid vm set status start fail when serivce restart (#24777) 2026-05-07 15:22:32 +08:00
Zexi Li
48aa05c5c7 Automated cherry pick of #24760: Feature/comfyui instant model (#24776) 
* feat(llm): support cursor

* feat(llm): support import from huggingface

* fix(llm): direction error

* feat(llm): support instant model for comfyui

* feat(llm): support multi file select when save-instant-model

---------

Co-authored-by: cwz <cwz_eikoh@163.com>
 2026-05-07 14:43:10 +08:00
wanyaoqi
3c919a5e71 feat(host): mps add container device configuration (#24773) 2026-05-07 10:03:31 +08:00
Jian Qiu
f998448ad1 fix: update vendor for v4.0.3 (#24761) 
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
v4.0.3-20260504.0 		2026-05-03 23:22:56 +08:00
Jian Qiu
71648a8e02 fix(host): cpu hotplug compatible with high qemu version (#24762) 
Co-authored-by: wanyaoqi <d3lx.yq@gmail.com>
 2026-05-03 23:22:31 +08:00
Jian Qiu
ece246f009 fix(host): auto merge backinf file generate disks index (#24763) 
Co-authored-by: wanyaoqi <d3lx.yq@gmail.com>
 2026-05-03 22:39:11 +08:00
Zexi Li
4eb015e649 fix(monitor): normalize resource-metrics alert_state (#24756) 
Treat attach and init as init in PerformResourceMetrics output.
v4.0.3-20260503.0 		2026-04-30 18:42:01 +08:00
Zexi Li
0005ea8229 fix(monitor): support influxql derivative and difference function translation (#24748) 2026-04-30 13:11:45 +08:00
Jian Qiu
225791f78c fix: init ifb kernel module (#24745) 
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
 2026-04-29 23:48:59 +08:00
wanyaoqi
e9dde8eae6 fix(mcclient,region): add tasks mod, service app options (#24734) 2026-04-29 23:48:06 +08:00
屈轩
5375fff061 fix(region): optimized listener rule list (#24742) 2026-04-29 14:09:29 +08:00
Zexi Li
0f30ccf520 fix(host,llm): fix pod shm mount path return and llm restart memory config (#24737) 
- Return shmPath instead of empty string when mountpoint is already
  mounted, so callers get the correct path
- Remove incorrect +1 offset in memory comparison and size formatting
  during LLM restart config change
 2026-04-28 19:57:42 +08:00
Zexi Li
a403ad5091 fix(lbagent,monitor): use victoria-metrics for telegraf and fix influxql translation (#24732) 2026-04-27 16:41:55 +08:00
Jian Qiu
b1908b24e3 fix: create redirect lb listener fail (#24725) 
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
 2026-04-27 10:01:19 +08:00
屈轩
acdf623027 fix(region): lb redirect rule create validate (#24729) 2026-04-27 10:00:54 +08:00
wanyaoqi
e42d517b28 chore(host-deployer): update host-deployer-base to 1.5.9 (#24722) 2026-04-26 23:02:59 +08:00
Zexi Li
9be35e4b45 feat(llm): support OpenClaw manual config mode (#24719) 
When ManualConfig is enabled, inject OPENCLAW_MANUAL_CONFIG=1 env var
and skip provider/channel credential injection, allowing users to
configure providers and channels themselves inside the desktop.
 2026-04-24 11:45:46 +08:00