Commit Graph
353 Commits
Author SHA1 Message Date
wanyaoqi 09785f2b71 feat(host): container add nvidia gpu share device type (#22432) 2025-04-17 16:56:19 +08:00
wanyaoqi 166698bdd2 feat(host): cgroupv2 support (#22314) 2025-03-28 16:13:05 +08:00
Jian QiuandQiu Jian a778fb9dd5 fix: host cache image directly from object storage (#22317)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2025-03-22 08:33:42 +08:00
wanyaoqi 44596e07dc fix(host): virtio nic add fdb record on has sriov nic (#22285) 2025-03-18 15:10:40 +08:00
wanyaoqi 91ae9fa50d fix(host): check image desc is empty (#22057)
image.inf record chksum or image id not matched will cause image desc empty
2025-02-05 18:53:58 +08:00
Jian QiuandQiu Jian 5a77b6bba5 fix: image cache use specific workers (#22046)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2025-02-05 11:33:28 +08:00
wanyaoqi a4377527c5 feat(host): container nvidia gpu metrics (#21950)
* feat(host): container nvidia gpu metrics

* feat(host): add vastaitech gpu metrics support

* feat(host): add cph amd gpu metrics

* fix(host): add nvidia gpu no card envs

* fix(host): add nvidia gpu telegraf conf on container host
2025-01-26 12:52:28 +08:00
wanyaoqi dbb86f8d30 feat(region,host): hostpinger add cpu usage percent report (#22005) 2025-01-25 15:42:45 +08:00
Zexi Li 085f9cca87 fix(host): netmask updated of hostlocal network (#21988) 2025-01-18 05:51:33 +08:00
Zexi Li 324cfe02f6 fix(host): wrap qemu machine info error (#21974) 2025-01-14 20:50:13 +08:00
wanyaoqi 2459378ace fix(region,host): guest reserve cpus compatible with numa allocate (#21803)
* fix(region,host): isolate guest reserved cpus

* fix(region,host): guest reserve cpus
2025-01-07 20:43:54 +08:00
Zexi Li a160b380c5 fix(host): make telegraf configuration log to debug (#21844) 2024-12-19 15:56:32 +08:00
Zexi Li 25a17bf338 feat(host): remove loop device by ioctl (#21676) 2024-12-05 10:12:49 +08:00
Zexi Li e2c244e2cb optimized(host): use goroutine to start ContainerStatsProvider (#21683) 2024-11-25 20:57:08 +08:00
wanyaoqi 84ecb2ec49 fix(region,host,scheduler): isolated device balance with to reserve cpus (#21614) 2024-11-17 22:45:59 +08:00
Zexi Li bdc9d8e289 feat(host,monitor): 支持容器相关gpu设备监控采集 (#21588) 2024-11-14 14:23:25 +08:00
Jian QiuandQiu Jian 50344061ea fix: integrate opentsdb (#21589)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2024-11-14 10:25:32 +08:00
wanyaoqi 63db9fc8e4 fix(scheduler): numa memory more balance on with prefer nodes (#21580)
* fix(scheduler): numa memory more balance on with prefer nodes

* fix(host): host agent allocate numa node
2024-11-14 10:18:08 +08:00
Jian QiuandQiu Jian dd848785ef fix: disable default dns domain cloud.onecloud.io (#21552)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2024-11-08 10:09:34 +08:00
Jian QiuandQiu Jian f42b4326a6 fix: telegraf kakfa ordered configuration (#21548)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2024-11-07 09:49:21 +08:00
wanyaoqi 0e7d28b76c fix(region,host): storage add master host (#21472) 2024-11-06 10:04:01 +08:00
wanyaoqi dc88d51d12 fix(region,scheduler,host): container numa aware (#21506)
- optimize host register isolated device update
- container device base try detect numa node
- container alloc numa memory
- calc numa distance before select isolated device

Signed-off-by: wanyaoqi <d3lx.yq@gmail.com>
2024-11-02 07:50:32 +08:00
Jian QiuandQiu Jian 3b396966dc feature: kafka telegraf output options (#21493)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2024-10-30 12:06:54 +08:00
Jian QiuandQiu Jian a8b8c89117 fix: host bond slave nic should use permaddr as mac (#21432)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2024-10-20 14:42:00 +08:00
wanyaoqi 331c5198d0 fix(host): set ssd disk default scheduler none (#21364) 2024-10-12 10:03:43 +08:00
Jian QiuandQiu Jian cd2e2f9020 fix: host report ovs kmod version and ksm status (#21330)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2024-09-28 20:23:56 +08:00
wanyaoqi 8ba1d316d9 fix(host): container custom cpufreq (#21285) 2024-09-25 10:29:33 +08:00
Zexi Li 8564725845 optimize(host): wrap updateOrCreateHost error (#21142) 2024-09-02 13:28:38 +08:00
Jian QiuandQiu Jian c451b7a2a1 fix: migrate ipv6 routes from physical interface to bridge (#21133)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2024-09-01 20:16:59 +08:00
Jian QiuandQiu Jian 077e7975f2 fix: host save host_local net confs in file (#21090)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2024-08-24 19:44:15 +08:00
Zexi Li ac05ef7b79 fix(monitor): container host alerting is not working (#21080) 2024-08-22 20:42:07 +08:00
Zexi Li d1e91ac543 optimize(region,host): add details error message when adding net interface (#21063) 2024-08-20 18:39:08 +08:00
Jian QiuandQiu Jian 61e0c4d3eb fix(host): migrate slave addresses and routes from physical nic to bridge (#21044)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2024-08-19 10:23:34 +08:00
Zexi Li addbcbe0d1 feat(region,host): startup probe of container (#21011) 2024-08-15 10:46:39 +08:00
Zexi Li 33da931b74 feat(host,region): support vastaitech gpu for container (#20987) 2024-08-12 09:33:50 +08:00
wanyaoqi b4861abe59 fix(host): isolated device probe closure reference for range val (#20988) 2024-08-09 16:17:55 +08:00
wanyaoqi 8d05cf2067 Fix/contaier alloc cpuset optimize sync isolated device (#20980)
* fix(region): optimize sync isolated devices

* fix(region,host): allocate cpuset for container
2024-08-09 09:39:47 +08:00
wanyaoqi 111b8ed678 feat(region,host,scheduler): scheduler support allocate cpu numa pin (#20932) 2024-07-30 19:10:38 +08:00
wanyaoqi e569326b83 feat(regin,host,host-deployer): qga deploy and fsdriver (#20864)
- host-deployer add deploy qga support
- host add qga partition fsdriver
- qga set network use fsdriver
- hostpinger sync qga status
2024-07-25 11:50:05 +08:00
Jian QiuandQiu Jian c8512e9da4 fix: always add static routes for metadata server (#20806)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2024-07-17 10:06:44 +08:00
Jian QiuandQiu Jian 44183fb021 feature: support baremetal profile (#20687)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2024-06-30 22:22:45 +08:00
wanyaoqi ddee083efa fix(region,host): isolated device batch sync (#20537) 2024-06-16 23:06:36 +08:00
wanyaoqi cf48918ca3 feat(region,host,scheduler): container ascend npu isolated device (#20411) 2024-05-31 00:11:16 +08:00
Jian QiuandQiu Jian 152b00d2d5 fix: support host local network (#20349)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2024-05-24 12:20:39 +08:00
Zexi Li b739cbd0dc feat(host): container metrics (#20257) 2024-05-16 16:00:29 +08:00
Jian QiuandQiu Jian ee943b6c31 feature: automatically clean obsolete images from host iamge cache (#20209)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2024-05-09 12:11:03 +08:00
wanyaoqi 5fc42614c3 feat(region,host): container mps support (#20138)
* feat(region,host): container mps support

* fix: gencopyright
2024-04-30 17:39:40 +08:00
wanyaoqi 7aae1d513a fix(host): detect sriov bond nics 2024-04-15 18:55:46 +08:00
Zexi Li ba46ca0219 feat(region,host,climc): simulating /sys/devices/system/cpu directory 2024-04-02 17:35:21 +08:00
Zexi Liandwanyaoqi 6313ce87a3 Support pod container (#19706)
* feat(region,scheduler,host): support pod and container

* feat(region,host): container nvidia gpu support

* fix(region,host): function parameters

---------

Co-authored-by: wanyaoqi <d3lx.yq@gmail.com>
2024-03-12 18:47:44 +08:00