Commit Graph

233 Commits

Author SHA1 Message Date
shenlan
c91cb3bb10 Merge pull request #42 from svc-design/codex/修改-gpu-k8s.sh-脚本
Update gpu install script
2025-07-02 17:13:56 +08:00
shenlan
8513984e49 Add gpu-k8s script 2025-07-02 17:12:28 +08:00
shenlan
844113d13e Merge pull request #41 from svc-design/codex/refactor-playbook-roles-into-directories
Add roles for k8s GPU and monitoring charts
2025-07-01 11:47:25 +08:00
shenlan
ae7a07d3d2 Add roles for GPU operator and monitoring charts 2025-07-01 11:47:11 +08:00
shenlan
9790a666da Merge pull request #40 from svc-design/codex/修复sealos_version变量未定义问题
Fix sealos_version fact when no GitHub response
2025-06-26 18:37:34 +08:00
shenlan
d6ffad1445 Fix undefined variable when fetching sealos release 2025-06-26 18:35:39 +08:00
Haitao Pan
3cd9f58488 Add GPU K8s reset playbook, rename init file, update inventory 2025-06-26 18:09:16 +08:00
shenlan
9c3defed67 Merge pull request #34 from svc-design/codex/rewrite-sealos-command-for-kubernetes-cluster
Add GPU k8s reset role and use script for sealos
2025-06-26 18:00:56 +08:00
shenlan
c080e71ac0 gpu-k8s role use script and add reset 2025-06-26 17:34:24 +08:00
Haitao Pan
7b0bea90e1 - fix(common): typo in add-apt-repository (universev -> universe -y)
- fix(gpu-k8s): fallback registry changed from ghcr.io/labring to labring
2025-06-26 14:39:37 +08:00
shenlan
5357f854f2 Merge pull request #32 from svc-design/codex/安装-fuse-overlayfs-并更新配置
Add fuse-overlayfs install note
2025-06-26 14:31:16 +08:00
shenlan
91f1231c50 Merge pull request #33 from svc-design/codex/修正镜像前缀错误导致403
Fix registry path concatenation
2025-06-26 13:50:35 +08:00
shenlan
0c89b74b73 fix: trim registry prefix when running sealos 2025-06-26 13:50:21 +08:00
Haitao Pan
af93640e0f fix: enable universe repo, use sudo for sealos, fix role order 2025-06-26 13:22:07 +08:00
shenlan
359bc52ef2 docs: add note on enabling universe 2025-06-26 13:14:45 +08:00
Haitao Pan
453fd924e7 fix: update node IPs and correct SSH key path for sealos 2025-06-26 13:06:03 +08:00
shenlan
2bbd367683 Install fuse-overlayfs and document requirement 2025-06-26 11:13:26 +08:00
shenlan
959579d9c0 Merge pull request #26 from svc-design/codex/修复sealos命令中未知标志错误
Fix Sealos ssh flags
2025-06-26 10:33:59 +08:00
shenlan
57be1121ea fix sealos ssh flags 2025-06-26 10:33:48 +08:00
shenlan
c724071a65 Merge pull request #25 from svc-design/codex/支持非-root-用户和-root-用户部署 2025-06-26 09:16:58 +08:00
shenlan
adddc2c186 Add non-root deployment support for gpu-k8s role 2025-06-26 09:16:11 +08:00
shenlan
12547b608c Merge pull request #24 from svc-design/codex/fix-ssh-connection-issue-to-ops-1
Allow customizing SSH user for cluster setup
2025-06-26 08:53:13 +08:00
shenlan
e810c963f4 Make ssh user configurable for cluster setup 2025-06-26 08:52:37 +08:00
shenlan
09e5059a63 Merge pull request #23 from svc-design/codex/分析并修复ssh无密码登录错误
Fix GPU k8s SSH trust
2025-06-26 00:07:14 +08:00
shenlan
f4e0e4fae2 Fix GPU cluster playbook SSH setup 2025-06-26 00:07:02 +08:00
shenlan
abea8f0727 Merge pull request #22 from svc-design/codex/fix-passwordless-ssh-access-issue
Fix GPU k8s ssh precheck user
2025-06-25 23:50:18 +08:00
shenlan
62c5a47591 fix(gpu-k8s): use inventory ssh user for precheck 2025-06-25 23:50:03 +08:00
shenlan
2bf9df3274 Merge pull request #21 from svc-design/codex/配置本机-ssh-key
Add SSH precheck for gpu-k8s role
2025-06-25 23:39:29 +08:00
shenlan
c5eaac2ce6 Authorize ops host key on all cluster nodes 2025-06-25 23:39:12 +08:00
shenlan
995725c793 gpu-k8s: precheck SSH connectivity 2025-06-25 23:29:49 +08:00
shenlan
098f2700fd Merge pull request #20 from svc-design/codex/fix-templating-error-in-ip-resolution
Fix IP resolution templating
2025-06-25 23:14:18 +08:00
shenlan
5e6a352286 Fix IP resolution templating 2025-06-25 23:14:00 +08:00
shenlan
caf844c486 Merge pull request #19 from svc-design/codex/修复---masters---nodes-未获取ip
Fix GPU Kubernetes IP resolution
2025-06-25 23:07:52 +08:00
shenlan
bee672a91a fix gpu-k8s role ip resolution 2025-06-25 23:07:35 +08:00
shenlan
b19d9f950b Merge pull request #18 from svc-design/codex/fix-permission-issue-with-get_labring_registry.sh
Fix GPU role variable checks
2025-06-25 22:56:06 +08:00
shenlan
fbf87ef5b7 support hostnames for gpu k8s role 2025-06-25 22:55:46 +08:00
shenlan
3cca3f04d1 Merge pull request #17 from svc-design/codex/fix--sudo--a-password-is-required--error
Fix LabRing registry prefix task sudo issue
2025-06-25 22:28:34 +08:00
shenlan
75ec257ff4 Move LabRing registry script into role 2025-06-25 22:27:46 +08:00
shenlan
f643a0bf4a Merge pull request #16 from svc-design/codex/修复labring注册表脚本未找到错误
Fix gpu-k8s role script path
2025-06-25 22:13:28 +08:00
shenlan
a0f553d89d fix gpu-k8s role script path 2025-06-25 22:13:14 +08:00
shenlan
b28e1e5ae8 Merge pull request #15 from svc-design/codex/根据节点ip选择镜像地址
Implement automatic LabRing registry selection
2025-06-25 22:05:50 +08:00
shenlan
856997c8eb Merge branch 'main' into codex/根据节点ip选择镜像地址 2025-06-25 22:04:13 +08:00
shenlan
e0b68c24d0 feat: auto-select labring registry 2025-06-25 21:47:00 +08:00
shenlan
2d296c95ac Merge pull request #14 from svc-design/codex/fix-invalid-ip-range-format-error
Fix GPU k8s role default version
2025-06-25 21:42:29 +08:00
shenlan
64eb61e38d gpu-k8s: separate kubernetes version 2025-06-25 21:42:15 +08:00
shenlan
6f65af587c Merge pull request #13 from svc-design/codex/fix-sealos-installation-404-error
Update gpu-k8s role to pull latest Sealos
2025-06-25 21:33:40 +08:00
shenlan
7164b20736 gpu role: fetch latest sealos and install tools 2025-06-25 21:33:22 +08:00
shenlan
b296acd33a Merge pull request #12 from svc-design/codex/修正roles/vhosts/gpu-k8s/配置与sealos初始化
Fix GPU K8S role and add ssh trust setup
2025-06-25 21:17:42 +08:00
shenlan
85be7e78e9 Add SSH trust role and enhance gpu-k8s setup 2025-06-25 21:17:30 +08:00
shenlan
e6c8bec317 Merge pull request #11 from svc-design/codex/更新-readme.md-并创建子文档
Update README with docs reference
2025-06-25 20:44:00 +08:00