skypilot-nightly 1.0.0.dev20251018__py3-none-any.whl → 1.0.0.dev20251021__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Potentially problematic release.

This version of skypilot-nightly might be problematic. Click here for more details.

Files changed (48) hide show
  1. sky/__init__.py +2 -2
  2. sky/adaptors/kubernetes.py +61 -0
  3. sky/backends/backend_utils.py +11 -11
  4. sky/backends/cloud_vm_ray_backend.py +15 -4
  5. sky/client/cli/command.py +39 -10
  6. sky/client/cli/flags.py +4 -2
  7. sky/client/sdk.py +26 -3
  8. sky/dashboard/out/404.html +1 -1
  9. sky/dashboard/out/clusters/[cluster]/[job].html +1 -1
  10. sky/dashboard/out/clusters/[cluster].html +1 -1
  11. sky/dashboard/out/clusters.html +1 -1
  12. sky/dashboard/out/config.html +1 -1
  13. sky/dashboard/out/index.html +1 -1
  14. sky/dashboard/out/infra/[context].html +1 -1
  15. sky/dashboard/out/infra.html +1 -1
  16. sky/dashboard/out/jobs/[job].html +1 -1
  17. sky/dashboard/out/jobs/pools/[pool].html +1 -1
  18. sky/dashboard/out/jobs.html +1 -1
  19. sky/dashboard/out/users.html +1 -1
  20. sky/dashboard/out/volumes.html +1 -1
  21. sky/dashboard/out/workspace/new.html +1 -1
  22. sky/dashboard/out/workspaces/[name].html +1 -1
  23. sky/dashboard/out/workspaces.html +1 -1
  24. sky/data/storage.py +2 -2
  25. sky/global_user_state.py +20 -20
  26. sky/jobs/server/server.py +10 -1
  27. sky/provision/kubernetes/network.py +9 -6
  28. sky/provision/provisioner.py +8 -0
  29. sky/serve/server/server.py +1 -0
  30. sky/server/common.py +9 -2
  31. sky/server/constants.py +1 -1
  32. sky/server/daemons.py +4 -2
  33. sky/server/requests/executor.py +10 -8
  34. sky/server/requests/payloads.py +2 -1
  35. sky/server/requests/preconditions.py +9 -4
  36. sky/server/requests/requests.py +124 -36
  37. sky/server/server.py +58 -24
  38. sky/server/stream_utils.py +127 -38
  39. sky/server/uvicorn.py +18 -17
  40. sky/utils/asyncio_utils.py +63 -3
  41. {skypilot_nightly-1.0.0.dev20251018.dist-info → skypilot_nightly-1.0.0.dev20251021.dist-info}/METADATA +33 -34
  42. {skypilot_nightly-1.0.0.dev20251018.dist-info → skypilot_nightly-1.0.0.dev20251021.dist-info}/RECORD +48 -48
  43. /sky/dashboard/out/_next/static/{Rn37hj-nuHOYT-HwxSDXC → jDc1PlRsl9Cc5FQUMLBu8}/_buildManifest.js +0 -0
  44. /sky/dashboard/out/_next/static/{Rn37hj-nuHOYT-HwxSDXC → jDc1PlRsl9Cc5FQUMLBu8}/_ssgManifest.js +0 -0
  45. {skypilot_nightly-1.0.0.dev20251018.dist-info → skypilot_nightly-1.0.0.dev20251021.dist-info}/WHEEL +0 -0
  46. {skypilot_nightly-1.0.0.dev20251018.dist-info → skypilot_nightly-1.0.0.dev20251021.dist-info}/entry_points.txt +0 -0
  47. {skypilot_nightly-1.0.0.dev20251018.dist-info → skypilot_nightly-1.0.0.dev20251021.dist-info}/licenses/LICENSE +0 -0
  48. {skypilot_nightly-1.0.0.dev20251018.dist-info → skypilot_nightly-1.0.0.dev20251021.dist-info}/top_level.txt +0 -0
sky/dashboard/out/jobs.html CHANGED
@@ -1 +1 @@
1
- <!DOCTYPE html><html><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width"/><meta name="next-head-count" content="2"/><link rel="preload" href="/dashboard/_next/static/css/4614e06482d7309e.css" as="style"/><link rel="stylesheet" href="/dashboard/_next/static/css/4614e06482d7309e.css" data-n-g=""/><noscript data-n-css=""></noscript><script defer="" nomodule="" src="/dashboard/_next/static/chunks/polyfills-78c92fac7aa8fdd8.js"></script><script src="/dashboard/_next/static/chunks/webpack-66f23594d38c7f16.js" defer=""></script><script src="/dashboard/_next/static/chunks/framework-cf60a09ccd051a10.js" defer=""></script><script src="/dashboard/_next/static/chunks/main-f15ccb73239a3bf1.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/_app-ce361c6959bc2001.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/jobs-a35a9dc3c5ccd657.js" defer=""></script><script src="/dashboard/_next/static/Rn37hj-nuHOYT-HwxSDXC/_buildManifest.js" defer=""></script><script src="/dashboard/_next/static/Rn37hj-nuHOYT-HwxSDXC/_ssgManifest.js" defer=""></script></head><body><div id="__next"></div><script id="__NEXT_DATA__" type="application/json">{"props":{"pageProps":{}},"page":"/jobs","query":{},"buildId":"Rn37hj-nuHOYT-HwxSDXC","assetPrefix":"/dashboard","nextExport":true,"autoExport":true,"isFallback":false,"scriptLoader":[]}</script></body></html>
1
+ <!DOCTYPE html><html><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width"/><meta name="next-head-count" content="2"/><link rel="preload" href="/dashboard/_next/static/css/4614e06482d7309e.css" as="style"/><link rel="stylesheet" href="/dashboard/_next/static/css/4614e06482d7309e.css" data-n-g=""/><noscript data-n-css=""></noscript><script defer="" nomodule="" src="/dashboard/_next/static/chunks/polyfills-78c92fac7aa8fdd8.js"></script><script src="/dashboard/_next/static/chunks/webpack-66f23594d38c7f16.js" defer=""></script><script src="/dashboard/_next/static/chunks/framework-cf60a09ccd051a10.js" defer=""></script><script src="/dashboard/_next/static/chunks/main-f15ccb73239a3bf1.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/_app-ce361c6959bc2001.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/jobs-a35a9dc3c5ccd657.js" defer=""></script><script src="/dashboard/_next/static/jDc1PlRsl9Cc5FQUMLBu8/_buildManifest.js" defer=""></script><script src="/dashboard/_next/static/jDc1PlRsl9Cc5FQUMLBu8/_ssgManifest.js" defer=""></script></head><body><div id="__next"></div><script id="__NEXT_DATA__" type="application/json">{"props":{"pageProps":{}},"page":"/jobs","query":{},"buildId":"jDc1PlRsl9Cc5FQUMLBu8","assetPrefix":"/dashboard","nextExport":true,"autoExport":true,"isFallback":false,"scriptLoader":[]}</script></body></html>
sky/dashboard/out/users.html CHANGED
@@ -1 +1 @@
1
- <!DOCTYPE html><html><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width"/><meta name="next-head-count" content="2"/><link rel="preload" href="/dashboard/_next/static/css/4614e06482d7309e.css" as="style"/><link rel="stylesheet" href="/dashboard/_next/static/css/4614e06482d7309e.css" data-n-g=""/><noscript data-n-css=""></noscript><script defer="" nomodule="" src="/dashboard/_next/static/chunks/polyfills-78c92fac7aa8fdd8.js"></script><script src="/dashboard/_next/static/chunks/webpack-66f23594d38c7f16.js" defer=""></script><script src="/dashboard/_next/static/chunks/framework-cf60a09ccd051a10.js" defer=""></script><script src="/dashboard/_next/static/chunks/main-f15ccb73239a3bf1.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/_app-ce361c6959bc2001.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/users-98d2ed979084162a.js" defer=""></script><script src="/dashboard/_next/static/Rn37hj-nuHOYT-HwxSDXC/_buildManifest.js" defer=""></script><script src="/dashboard/_next/static/Rn37hj-nuHOYT-HwxSDXC/_ssgManifest.js" defer=""></script></head><body><div id="__next"></div><script id="__NEXT_DATA__" type="application/json">{"props":{"pageProps":{}},"page":"/users","query":{},"buildId":"Rn37hj-nuHOYT-HwxSDXC","assetPrefix":"/dashboard","nextExport":true,"autoExport":true,"isFallback":false,"scriptLoader":[]}</script></body></html>
1
+ <!DOCTYPE html><html><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width"/><meta name="next-head-count" content="2"/><link rel="preload" href="/dashboard/_next/static/css/4614e06482d7309e.css" as="style"/><link rel="stylesheet" href="/dashboard/_next/static/css/4614e06482d7309e.css" data-n-g=""/><noscript data-n-css=""></noscript><script defer="" nomodule="" src="/dashboard/_next/static/chunks/polyfills-78c92fac7aa8fdd8.js"></script><script src="/dashboard/_next/static/chunks/webpack-66f23594d38c7f16.js" defer=""></script><script src="/dashboard/_next/static/chunks/framework-cf60a09ccd051a10.js" defer=""></script><script src="/dashboard/_next/static/chunks/main-f15ccb73239a3bf1.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/_app-ce361c6959bc2001.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/users-98d2ed979084162a.js" defer=""></script><script src="/dashboard/_next/static/jDc1PlRsl9Cc5FQUMLBu8/_buildManifest.js" defer=""></script><script src="/dashboard/_next/static/jDc1PlRsl9Cc5FQUMLBu8/_ssgManifest.js" defer=""></script></head><body><div id="__next"></div><script id="__NEXT_DATA__" type="application/json">{"props":{"pageProps":{}},"page":"/users","query":{},"buildId":"jDc1PlRsl9Cc5FQUMLBu8","assetPrefix":"/dashboard","nextExport":true,"autoExport":true,"isFallback":false,"scriptLoader":[]}</script></body></html>
sky/dashboard/out/volumes.html CHANGED
@@ -1 +1 @@
1
- <!DOCTYPE html><html><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width"/><meta name="next-head-count" content="2"/><link rel="preload" href="/dashboard/_next/static/css/4614e06482d7309e.css" as="style"/><link rel="stylesheet" href="/dashboard/_next/static/css/4614e06482d7309e.css" data-n-g=""/><noscript data-n-css=""></noscript><script defer="" nomodule="" src="/dashboard/_next/static/chunks/polyfills-78c92fac7aa8fdd8.js"></script><script src="/dashboard/_next/static/chunks/webpack-66f23594d38c7f16.js" defer=""></script><script src="/dashboard/_next/static/chunks/framework-cf60a09ccd051a10.js" defer=""></script><script src="/dashboard/_next/static/chunks/main-f15ccb73239a3bf1.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/_app-ce361c6959bc2001.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/volumes-835d14ba94808f79.js" defer=""></script><script src="/dashboard/_next/static/Rn37hj-nuHOYT-HwxSDXC/_buildManifest.js" defer=""></script><script src="/dashboard/_next/static/Rn37hj-nuHOYT-HwxSDXC/_ssgManifest.js" defer=""></script></head><body><div id="__next"></div><script id="__NEXT_DATA__" type="application/json">{"props":{"pageProps":{}},"page":"/volumes","query":{},"buildId":"Rn37hj-nuHOYT-HwxSDXC","assetPrefix":"/dashboard","nextExport":true,"autoExport":true,"isFallback":false,"scriptLoader":[]}</script></body></html>
1
+ <!DOCTYPE html><html><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width"/><meta name="next-head-count" content="2"/><link rel="preload" href="/dashboard/_next/static/css/4614e06482d7309e.css" as="style"/><link rel="stylesheet" href="/dashboard/_next/static/css/4614e06482d7309e.css" data-n-g=""/><noscript data-n-css=""></noscript><script defer="" nomodule="" src="/dashboard/_next/static/chunks/polyfills-78c92fac7aa8fdd8.js"></script><script src="/dashboard/_next/static/chunks/webpack-66f23594d38c7f16.js" defer=""></script><script src="/dashboard/_next/static/chunks/framework-cf60a09ccd051a10.js" defer=""></script><script src="/dashboard/_next/static/chunks/main-f15ccb73239a3bf1.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/_app-ce361c6959bc2001.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/volumes-835d14ba94808f79.js" defer=""></script><script src="/dashboard/_next/static/jDc1PlRsl9Cc5FQUMLBu8/_buildManifest.js" defer=""></script><script src="/dashboard/_next/static/jDc1PlRsl9Cc5FQUMLBu8/_ssgManifest.js" defer=""></script></head><body><div id="__next"></div><script id="__NEXT_DATA__" type="application/json">{"props":{"pageProps":{}},"page":"/volumes","query":{},"buildId":"jDc1PlRsl9Cc5FQUMLBu8","assetPrefix":"/dashboard","nextExport":true,"autoExport":true,"isFallback":false,"scriptLoader":[]}</script></body></html>
sky/dashboard/out/workspace/new.html CHANGED
@@ -1 +1 @@
1
- <!DOCTYPE html><html><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width"/><meta name="next-head-count" content="2"/><link rel="preload" href="/dashboard/_next/static/css/4614e06482d7309e.css" as="style"/><link rel="stylesheet" href="/dashboard/_next/static/css/4614e06482d7309e.css" data-n-g=""/><noscript data-n-css=""></noscript><script defer="" nomodule="" src="/dashboard/_next/static/chunks/polyfills-78c92fac7aa8fdd8.js"></script><script src="/dashboard/_next/static/chunks/webpack-66f23594d38c7f16.js" defer=""></script><script src="/dashboard/_next/static/chunks/framework-cf60a09ccd051a10.js" defer=""></script><script src="/dashboard/_next/static/chunks/main-f15ccb73239a3bf1.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/_app-ce361c6959bc2001.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/workspace/new-3f88a1c7e86a3f86.js" defer=""></script><script src="/dashboard/_next/static/Rn37hj-nuHOYT-HwxSDXC/_buildManifest.js" defer=""></script><script src="/dashboard/_next/static/Rn37hj-nuHOYT-HwxSDXC/_ssgManifest.js" defer=""></script></head><body><div id="__next"></div><script id="__NEXT_DATA__" type="application/json">{"props":{"pageProps":{}},"page":"/workspace/new","query":{},"buildId":"Rn37hj-nuHOYT-HwxSDXC","assetPrefix":"/dashboard","nextExport":true,"autoExport":true,"isFallback":false,"scriptLoader":[]}</script></body></html>
1
+ <!DOCTYPE html><html><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width"/><meta name="next-head-count" content="2"/><link rel="preload" href="/dashboard/_next/static/css/4614e06482d7309e.css" as="style"/><link rel="stylesheet" href="/dashboard/_next/static/css/4614e06482d7309e.css" data-n-g=""/><noscript data-n-css=""></noscript><script defer="" nomodule="" src="/dashboard/_next/static/chunks/polyfills-78c92fac7aa8fdd8.js"></script><script src="/dashboard/_next/static/chunks/webpack-66f23594d38c7f16.js" defer=""></script><script src="/dashboard/_next/static/chunks/framework-cf60a09ccd051a10.js" defer=""></script><script src="/dashboard/_next/static/chunks/main-f15ccb73239a3bf1.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/_app-ce361c6959bc2001.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/workspace/new-3f88a1c7e86a3f86.js" defer=""></script><script src="/dashboard/_next/static/jDc1PlRsl9Cc5FQUMLBu8/_buildManifest.js" defer=""></script><script src="/dashboard/_next/static/jDc1PlRsl9Cc5FQUMLBu8/_ssgManifest.js" defer=""></script></head><body><div id="__next"></div><script id="__NEXT_DATA__" type="application/json">{"props":{"pageProps":{}},"page":"/workspace/new","query":{},"buildId":"jDc1PlRsl9Cc5FQUMLBu8","assetPrefix":"/dashboard","nextExport":true,"autoExport":true,"isFallback":false,"scriptLoader":[]}</script></body></html>
sky/dashboard/out/workspaces/[name].html CHANGED
@@ -1 +1 @@
1
- <!DOCTYPE html><html><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width"/><meta name="next-head-count" content="2"/><link rel="preload" href="/dashboard/_next/static/css/4614e06482d7309e.css" as="style"/><link rel="stylesheet" href="/dashboard/_next/static/css/4614e06482d7309e.css" data-n-g=""/><noscript data-n-css=""></noscript><script defer="" nomodule="" src="/dashboard/_next/static/chunks/polyfills-78c92fac7aa8fdd8.js"></script><script src="/dashboard/_next/static/chunks/webpack-66f23594d38c7f16.js" defer=""></script><script src="/dashboard/_next/static/chunks/framework-cf60a09ccd051a10.js" defer=""></script><script src="/dashboard/_next/static/chunks/main-f15ccb73239a3bf1.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/_app-ce361c6959bc2001.js" defer=""></script><script src="/dashboard/_next/static/chunks/616-3d59f75e2ccf9321.js" defer=""></script><script src="/dashboard/_next/static/chunks/6130-2be46d70a38f1e82.js" defer=""></script><script src="/dashboard/_next/static/chunks/5739-d67458fcb1386c92.js" defer=""></script><script src="/dashboard/_next/static/chunks/7411-b15471acd2cba716.js" defer=""></script><script src="/dashboard/_next/static/chunks/1272-1ef0bf0237faccdb.js" defer=""></script><script src="/dashboard/_next/static/chunks/7359-c8d04e06886000b3.js" defer=""></script><script src="/dashboard/_next/static/chunks/6989-01359c57e018caa4.js" defer=""></script><script src="/dashboard/_next/static/chunks/3850-ff4a9a69d978632b.js" defer=""></script><script src="/dashboard/_next/static/chunks/8969-66237729cdf9749e.js" defer=""></script><script src="/dashboard/_next/static/chunks/6990-f6818c84ed8f1c86.js" defer=""></script><script src="/dashboard/_next/static/chunks/6135-4b4d5e824b7f9d3c.js" defer=""></script><script src="/dashboard/_next/static/chunks/1121-d0782b9251f0fcd3.js" defer=""></script><script src="/dashboard/_next/static/chunks/6601-06114c982db410b6.js" defer=""></script><script src="/dashboard/_next/static/chunks/3015-7e0e8f06bb2f881c.js" defer=""></script><script src="/dashboard/_next/static/chunks/1141-3b40c39626f99c89.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/workspaces/%5Bname%5D-e8688c35c06f0ac5.js" defer=""></script><script src="/dashboard/_next/static/Rn37hj-nuHOYT-HwxSDXC/_buildManifest.js" defer=""></script><script src="/dashboard/_next/static/Rn37hj-nuHOYT-HwxSDXC/_ssgManifest.js" defer=""></script></head><body><div id="__next"></div><script id="__NEXT_DATA__" type="application/json">{"props":{"pageProps":{}},"page":"/workspaces/[name]","query":{},"buildId":"Rn37hj-nuHOYT-HwxSDXC","assetPrefix":"/dashboard","nextExport":true,"autoExport":true,"isFallback":false,"scriptLoader":[]}</script></body></html>
1
+ <!DOCTYPE html><html><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width"/><meta name="next-head-count" content="2"/><link rel="preload" href="/dashboard/_next/static/css/4614e06482d7309e.css" as="style"/><link rel="stylesheet" href="/dashboard/_next/static/css/4614e06482d7309e.css" data-n-g=""/><noscript data-n-css=""></noscript><script defer="" nomodule="" src="/dashboard/_next/static/chunks/polyfills-78c92fac7aa8fdd8.js"></script><script src="/dashboard/_next/static/chunks/webpack-66f23594d38c7f16.js" defer=""></script><script src="/dashboard/_next/static/chunks/framework-cf60a09ccd051a10.js" defer=""></script><script src="/dashboard/_next/static/chunks/main-f15ccb73239a3bf1.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/_app-ce361c6959bc2001.js" defer=""></script><script src="/dashboard/_next/static/chunks/616-3d59f75e2ccf9321.js" defer=""></script><script src="/dashboard/_next/static/chunks/6130-2be46d70a38f1e82.js" defer=""></script><script src="/dashboard/_next/static/chunks/5739-d67458fcb1386c92.js" defer=""></script><script src="/dashboard/_next/static/chunks/7411-b15471acd2cba716.js" defer=""></script><script src="/dashboard/_next/static/chunks/1272-1ef0bf0237faccdb.js" defer=""></script><script src="/dashboard/_next/static/chunks/7359-c8d04e06886000b3.js" defer=""></script><script src="/dashboard/_next/static/chunks/6989-01359c57e018caa4.js" defer=""></script><script src="/dashboard/_next/static/chunks/3850-ff4a9a69d978632b.js" defer=""></script><script src="/dashboard/_next/static/chunks/8969-66237729cdf9749e.js" defer=""></script><script src="/dashboard/_next/static/chunks/6990-f6818c84ed8f1c86.js" defer=""></script><script src="/dashboard/_next/static/chunks/6135-4b4d5e824b7f9d3c.js" defer=""></script><script src="/dashboard/_next/static/chunks/1121-d0782b9251f0fcd3.js" defer=""></script><script src="/dashboard/_next/static/chunks/6601-06114c982db410b6.js" defer=""></script><script src="/dashboard/_next/static/chunks/3015-7e0e8f06bb2f881c.js" defer=""></script><script src="/dashboard/_next/static/chunks/1141-3b40c39626f99c89.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/workspaces/%5Bname%5D-e8688c35c06f0ac5.js" defer=""></script><script src="/dashboard/_next/static/jDc1PlRsl9Cc5FQUMLBu8/_buildManifest.js" defer=""></script><script src="/dashboard/_next/static/jDc1PlRsl9Cc5FQUMLBu8/_ssgManifest.js" defer=""></script></head><body><div id="__next"></div><script id="__NEXT_DATA__" type="application/json">{"props":{"pageProps":{}},"page":"/workspaces/[name]","query":{},"buildId":"jDc1PlRsl9Cc5FQUMLBu8","assetPrefix":"/dashboard","nextExport":true,"autoExport":true,"isFallback":false,"scriptLoader":[]}</script></body></html>
sky/dashboard/out/workspaces.html CHANGED
@@ -1 +1 @@
1
- <!DOCTYPE html><html><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width"/><meta name="next-head-count" content="2"/><link rel="preload" href="/dashboard/_next/static/css/4614e06482d7309e.css" as="style"/><link rel="stylesheet" href="/dashboard/_next/static/css/4614e06482d7309e.css" data-n-g=""/><noscript data-n-css=""></noscript><script defer="" nomodule="" src="/dashboard/_next/static/chunks/polyfills-78c92fac7aa8fdd8.js"></script><script src="/dashboard/_next/static/chunks/webpack-66f23594d38c7f16.js" defer=""></script><script src="/dashboard/_next/static/chunks/framework-cf60a09ccd051a10.js" defer=""></script><script src="/dashboard/_next/static/chunks/main-f15ccb73239a3bf1.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/_app-ce361c6959bc2001.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/workspaces-69c80d677d3c2949.js" defer=""></script><script src="/dashboard/_next/static/Rn37hj-nuHOYT-HwxSDXC/_buildManifest.js" defer=""></script><script src="/dashboard/_next/static/Rn37hj-nuHOYT-HwxSDXC/_ssgManifest.js" defer=""></script></head><body><div id="__next"></div><script id="__NEXT_DATA__" type="application/json">{"props":{"pageProps":{}},"page":"/workspaces","query":{},"buildId":"Rn37hj-nuHOYT-HwxSDXC","assetPrefix":"/dashboard","nextExport":true,"autoExport":true,"isFallback":false,"scriptLoader":[]}</script></body></html>
1
+ <!DOCTYPE html><html><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width"/><meta name="next-head-count" content="2"/><link rel="preload" href="/dashboard/_next/static/css/4614e06482d7309e.css" as="style"/><link rel="stylesheet" href="/dashboard/_next/static/css/4614e06482d7309e.css" data-n-g=""/><noscript data-n-css=""></noscript><script defer="" nomodule="" src="/dashboard/_next/static/chunks/polyfills-78c92fac7aa8fdd8.js"></script><script src="/dashboard/_next/static/chunks/webpack-66f23594d38c7f16.js" defer=""></script><script src="/dashboard/_next/static/chunks/framework-cf60a09ccd051a10.js" defer=""></script><script src="/dashboard/_next/static/chunks/main-f15ccb73239a3bf1.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/_app-ce361c6959bc2001.js" defer=""></script><script src="/dashboard/_next/static/chunks/pages/workspaces-69c80d677d3c2949.js" defer=""></script><script src="/dashboard/_next/static/jDc1PlRsl9Cc5FQUMLBu8/_buildManifest.js" defer=""></script><script src="/dashboard/_next/static/jDc1PlRsl9Cc5FQUMLBu8/_ssgManifest.js" defer=""></script></head><body><div id="__next"></div><script id="__NEXT_DATA__" type="application/json">{"props":{"pageProps":{}},"page":"/workspaces","query":{},"buildId":"jDc1PlRsl9Cc5FQUMLBu8","assetPrefix":"/dashboard","nextExport":true,"autoExport":true,"isFallback":false,"scriptLoader":[]}</script></body></html>
sky/data/storage.py CHANGED
@@ -2015,7 +2015,7 @@ class S3CompatibleStore(AbstractStore):
2015
2015
  except aws.botocore_exceptions().ClientError as e:
2016
2016
  with ux_utils.print_exception_no_traceback():
2017
2017
  raise exceptions.StorageBucketCreateError(
2018
- f'Attempted to create a bucket {self.name} but failed.'
2018
+ f'Attempted to create S3 bucket {self.name} but failed.'
2019
2019
  ) from e
2020
2020
  return self.config.resource_factory(bucket_name)
2021
2021
 
@@ -2554,7 +2554,7 @@ class GcsStore(AbstractStore):
2554
2554
  except Exception as e: # pylint: disable=broad-except
2555
2555
  with ux_utils.print_exception_no_traceback():
2556
2556
  raise exceptions.StorageBucketCreateError(
2557
- f'Attempted to create a bucket {self.name} but failed.'
2557
+ f'Attempted to create GCS bucket {self.name} but failed.'
2558
2558
  ) from e
2559
2559
  logger.info(
2560
2560
  f' {colorama.Style.DIM}Created GCS bucket {new_bucket.name!r} in '
sky/global_user_state.py CHANGED
@@ -1620,8 +1620,10 @@ def get_clusters(
1620
1620
  cluster_table.c.storage_mounts_metadata,
1621
1621
  cluster_table.c.cluster_ever_up,
1622
1622
  cluster_table.c.status_updated_at, cluster_table.c.user_hash,
1623
- cluster_table.c.config_hash, cluster_table.c.workspace,
1624
- cluster_table.c.is_managed)
1623
+ cluster_table.c.config_hash,
1624
+ cluster_table.c.workspace, cluster_table.c.is_managed,
1625
+ user_table.c.name.label('user_name')).outerjoin(
1626
+ user_table, cluster_table.c.user_hash == user_table.c.id)
1625
1627
  else:
1626
1628
  query = session.query(
1627
1629
  cluster_table.c.name,
@@ -1643,7 +1645,9 @@ def get_clusters(
1643
1645
  cluster_table.c.is_managed,
1644
1646
  # extra fields compared to above query
1645
1647
  cluster_table.c.last_creation_yaml,
1646
- cluster_table.c.last_creation_command)
1648
+ cluster_table.c.last_creation_command,
1649
+ user_table.c.name.label('user_name')).outerjoin(
1650
+ user_table, cluster_table.c.user_hash == user_table.c.id)
1647
1651
  if exclude_managed_clusters:
1648
1652
  query = query.filter(cluster_table.c.is_managed == int(False))
1649
1653
  if workspaces_filter is not None:
@@ -1666,28 +1670,22 @@ def get_clusters(
1666
1670
  rows = query.all()
1667
1671
  records = []
1668
1672
 
1669
- # get user hash for each row
1670
- row_to_user_hash = {}
1671
- for row in rows:
1672
- user_hash = (row.user_hash
1673
- if row.user_hash is not None else current_user_hash)
1674
- row_to_user_hash[row.cluster_hash] = user_hash
1675
-
1676
- # get all users needed for the rows at once
1677
- user_hashes = set(row_to_user_hash.values())
1678
- user_hash_to_user = get_users(user_hashes)
1673
+ # Check if we need to fetch the current user's name,
1674
+ # for backwards compatibility, if user_hash is None.
1675
+ current_user_name = None
1676
+ needs_current_user = any(row.user_hash is None for row in rows)
1677
+ if needs_current_user:
1678
+ current_user = get_user(current_user_hash)
1679
+ current_user_name = (current_user.name
1680
+ if current_user is not None else None)
1679
1681
 
1680
1682
  # get last cluster event for each row
1681
- cluster_hashes = set(row_to_user_hash.keys())
1682
1683
  if not summary_response:
1684
+ cluster_hashes = {row.cluster_hash for row in rows}
1683
1685
  last_cluster_event_dict = _get_last_cluster_event_multiple(
1684
1686
  cluster_hashes, ClusterEventType.STATUS_CHANGE)
1685
1687
 
1686
- # get user for each row
1687
1688
  for row in rows:
1688
- user_hash = row_to_user_hash[row.cluster_hash]
1689
- user = user_hash_to_user.get(user_hash, None)
1690
- user_name = user.name if user is not None else None
1691
1689
  # TODO: use namedtuple instead of dict
1692
1690
  record = {
1693
1691
  'name': row.name,
@@ -1704,8 +1702,10 @@ def get_clusters(
1704
1702
  row.storage_mounts_metadata),
1705
1703
  'cluster_ever_up': bool(row.cluster_ever_up),
1706
1704
  'status_updated_at': row.status_updated_at,
1707
- 'user_hash': user_hash,
1708
- 'user_name': user_name,
1705
+ 'user_hash': (row.user_hash
1706
+ if row.user_hash is not None else current_user_hash),
1707
+ 'user_name': (row.user_name
1708
+ if row.user_name is not None else current_user_name),
1709
1709
  'workspace': row.workspace,
1710
1710
  'is_managed': bool(row.is_managed),
1711
1711
  'config_hash': row.config_hash,
sky/jobs/server/server.py CHANGED
@@ -109,6 +109,7 @@ async def logs(
109
109
  schedule_type=schedule_type,
110
110
  request_cluster_name=common.JOB_CONTROLLER_NAME,
111
111
  )
112
+ kill_request_on_disconnect = False
112
113
  if schedule_type == api_requests.ScheduleType.SHORT:
113
114
  # For short request, run in the coroutine to avoid blocking
114
115
  # short workers.
@@ -117,11 +118,15 @@ async def logs(
117
118
  background_tasks.add_task(task.cancel)
118
119
  else:
119
120
  executor.schedule_prepared_request(request_task)
121
+ # When runs in long executor process, we should kill the request on
122
+ # disconnect to cancel the running routine.
123
+ kill_request_on_disconnect = True
120
124
 
121
125
  return stream_utils.stream_response_for_long_request(
122
126
  request_id=request_task.request_id,
123
127
  logs_path=request_task.log_path,
124
128
  background_tasks=background_tasks,
129
+ kill_request_on_disconnect=kill_request_on_disconnect,
125
130
  )
126
131
 
127
132
 
@@ -201,12 +206,16 @@ async def pool_tail_logs(
201
206
  request_cluster_name=common.JOB_CONTROLLER_NAME,
202
207
  )
203
208
 
204
- request_task = api_requests.get_request(request.state.request_id)
209
+ request_task = api_requests.get_request(request.state.request_id,
210
+ fields=['request_id'])
205
211
 
206
212
  return stream_utils.stream_response_for_long_request(
207
213
  request_id=request_task.request_id,
214
+ # req.log_path is derived from request_id,
215
+ # so it's ok to just grab the request_id in the above query.
208
216
  logs_path=request_task.log_path,
209
217
  background_tasks=background_tasks,
218
+ kill_request_on_disconnect=True,
210
219
  )
211
220
 
212
221
 
sky/provision/kubernetes/network.py CHANGED
@@ -48,8 +48,10 @@ def _open_ports_using_loadbalancer(
48
48
  service_name = _LOADBALANCER_SERVICE_NAME.format(
49
49
  cluster_name_on_cloud=cluster_name_on_cloud)
50
50
  context = kubernetes_utils.get_context_from_config(provider_config)
51
+ namespace = kubernetes_utils.get_namespace_from_config(provider_config)
52
+
51
53
  content = network_utils.fill_loadbalancer_template(
52
- namespace=provider_config.get('namespace', 'default'),
54
+ namespace=namespace,
53
55
  context=context,
54
56
  service_name=service_name,
55
57
  ports=ports,
@@ -103,7 +105,7 @@ def _open_ports_using_ingress(
103
105
  # To avoid this, we change ingress creation into one object containing
104
106
  # multiple rules.
105
107
  content = network_utils.fill_ingress_template(
106
- namespace=provider_config.get('namespace', 'default'),
108
+ namespace=namespace,
107
109
  context=context,
108
110
  service_details=service_details,
109
111
  ingress_name=f'{cluster_name_on_cloud}-skypilot-ingress',
@@ -165,9 +167,10 @@ def _cleanup_ports_for_loadbalancer(
165
167
  # TODO(aylei): test coverage
166
168
  context = provider_config.get(
167
169
  'context', kubernetes_utils.get_current_kube_config_context_name())
170
+ namespace = kubernetes_utils.get_namespace_from_config(provider_config)
168
171
  network_utils.delete_namespaced_service(
169
172
  context=context,
170
- namespace=provider_config.get('namespace', 'default'),
173
+ namespace=namespace,
171
174
  service_name=service_name,
172
175
  )
173
176
 
@@ -180,19 +183,19 @@ def _cleanup_ports_for_ingress(
180
183
  # Delete services for each port
181
184
  context = provider_config.get(
182
185
  'context', kubernetes_utils.get_current_kube_config_context_name())
186
+ namespace = kubernetes_utils.get_namespace_from_config(provider_config)
183
187
  for port in ports:
184
188
  service_name = f'{cluster_name_on_cloud}--skypilot-svc--{port}'
185
189
  network_utils.delete_namespaced_service(
186
190
  context=context,
187
- namespace=provider_config.get('namespace',
188
- kubernetes_utils.DEFAULT_NAMESPACE),
191
+ namespace=namespace,
189
192
  service_name=service_name,
190
193
  )
191
194
 
192
195
  # Delete the single ingress used for all ports
193
196
  ingress_name = f'{cluster_name_on_cloud}-skypilot-ingress'
194
197
  network_utils.delete_namespaced_ingress(
195
- namespace=kubernetes_utils.get_namespace_from_config(provider_config),
198
+ namespace=namespace,
196
199
  context=kubernetes_utils.get_context_from_config(provider_config),
197
200
  ingress_name=ingress_name,
198
201
  )
sky/provision/provisioner.py CHANGED
@@ -442,6 +442,14 @@ def _post_provision_setup(
442
442
  cluster_name.name_on_cloud,
443
443
  provider_config=provider_config)
444
444
 
445
+ # Update cluster info in handle so cluster instance ids are set. This
446
+ # allows us to expose provision logs to debug nodes that failed during post
447
+ # provision setup.
448
+ handle = global_user_state.get_handle_from_cluster_name(
449
+ cluster_name.display_name)
450
+ handle.cached_cluster_info = cluster_info
451
+ global_user_state.update_cluster_handle(cluster_name.display_name, handle)
452
+
445
453
  if cluster_info.num_instances > 1:
446
454
  # Only worker nodes have logs in the per-instance log directory. Head
447
455
  # node's log will be redirected to the main log file.
sky/serve/server/server.py CHANGED
@@ -114,6 +114,7 @@ async def tail_logs(
114
114
  request_id=request_task.request_id,
115
115
  logs_path=request_task.log_path,
116
116
  background_tasks=background_tasks,
117
+ kill_request_on_disconnect=False,
117
118
  )
118
119
 
119
120
 
sky/server/common.py CHANGED
@@ -490,6 +490,7 @@ def get_api_server_status(endpoint: Optional[str] = None) -> ApiServerInfo:
490
490
  def handle_request_error(response: 'requests.Response') -> None:
491
491
  # Keep the original HTTPError if the response code >= 400
492
492
  response.raise_for_status()
493
+
493
494
  # Other status codes are not expected neither, e.g. we do not expect to
494
495
  # handle redirection here.
495
496
  if response.status_code != 200:
@@ -916,12 +917,18 @@ def reload_for_new_request(client_entrypoint: Optional[str],
916
917
  client_command: Optional[str],
917
918
  using_remote_api_server: bool, user: 'models.User',
918
919
  request_id: str) -> None:
919
- """Reload modules, global variables, and usage message for a new request."""
920
+ """Reload modules, global variables, and usage message for a new request.
921
+
922
+ Must be called within the request's context.
923
+ """
920
924
  # This should be called first to make sure the logger is up-to-date.
921
925
  sky_logging.reload_logger()
922
926
 
923
927
  # Reload the skypilot config to make sure the latest config is used.
924
- skypilot_config.safe_reload_config()
928
+ # We don't need to grab the lock here because this function is only
929
+ # run once we are inside the request's context, so there shouldn't
930
+ # be any race conditions when reloading the config.
931
+ skypilot_config.reload_config()
925
932
 
926
933
  # Reset the client entrypoint and command for the usage message.
927
934
  common_utils.set_request_context(
sky/server/constants.py CHANGED
@@ -10,7 +10,7 @@ from sky.skylet import constants
10
10
  # based on version info is needed.
11
11
  # For more details and code guidelines, refer to:
12
12
  # https://docs.skypilot.co/en/latest/developers/CONTRIBUTING.html#backward-compatibility-guidelines
13
- API_VERSION = 20
13
+ API_VERSION = 21
14
14
 
15
15
  # The minimum peer API version that the code should still work with.
16
16
  # Notes (dev):
sky/server/daemons.py CHANGED
@@ -38,9 +38,11 @@ class InternalRequestDaemon:
38
38
  try:
39
39
  # Refresh config within the while loop.
40
40
  # Since this is a long running daemon,
41
- # reload_config_for_new_request()
41
+ # reload_for_new_request()
42
42
  # is not called in between the event runs.
43
- skypilot_config.safe_reload_config()
43
+ # We don't need to grab the lock here because each of the daemons
44
+ # run in their own process and thus have their own request context.
45
+ skypilot_config.reload_config()
44
46
  # Get the configured log level for the daemon inside the event loop
45
47
  # in case the log level changes after the API server is started.
46
48
  level_str = skypilot_config.get_nested(
sky/server/requests/executor.py CHANGED
@@ -214,10 +214,11 @@ class RequestWorker:
214
214
  time.sleep(0.1)
215
215
  return
216
216
  request_id, ignore_return_value, _ = request_element
217
- request = api_requests.get_request(request_id)
217
+ request = api_requests.get_request(request_id, fields=['status'])
218
218
  assert request is not None, f'Request with ID {request_id} is None'
219
219
  if request.status == api_requests.RequestStatus.CANCELLED:
220
220
  return
221
+ del request
221
222
  logger.info(f'[{self}] Submitting request: {request_id}')
222
223
  # Start additional process to run the request, so that it can be
223
224
  # cancelled when requested by a user.
@@ -621,8 +622,8 @@ async def _execute_request_coroutine(request: api_requests.Request):
621
622
  logger.info(f'Executing request {request.request_id} in coroutine')
622
623
  func = request.entrypoint
623
624
  request_body = request.request_body
624
- with api_requests.update_request(request.request_id) as request_task:
625
- request_task.status = api_requests.RequestStatus.RUNNING
625
+ await api_requests.update_status_async(request.request_id,
626
+ api_requests.RequestStatus.RUNNING)
626
627
  # Redirect stdout and stderr to the request log path.
627
628
  original_output = ctx.redirect_log(request.log_path)
628
629
  try:
@@ -632,7 +633,7 @@ async def _execute_request_coroutine(request: api_requests.Request):
632
633
  **request_body.to_kwargs())
633
634
  except Exception as e: # pylint: disable=broad-except
634
635
  ctx.redirect_log(original_output)
635
- api_requests.set_request_failed(request.request_id, e)
636
+ await api_requests.set_request_failed_async(request.request_id, e)
636
637
  logger.error(f'Failed to run request {request.request_id} due to '
637
638
  f'{common_utils.format_exception(e)}')
638
639
  return
@@ -649,14 +650,15 @@ async def _execute_request_coroutine(request: api_requests.Request):
649
650
  if fut.done():
650
651
  try:
651
652
  result = await fut
652
- api_requests.set_request_succeeded(request_id, result)
653
+ await api_requests.set_request_succeeded_async(
654
+ request_id, result)
653
655
  except asyncio.CancelledError:
654
656
  # The task is cancelled by ctx.cancel(), where the status
655
657
  # should already be set to CANCELLED.
656
658
  pass
657
659
  except Exception as e: # pylint: disable=broad-except
658
660
  ctx.redirect_log(original_output)
659
- api_requests.set_request_failed(request_id, e)
661
+ await api_requests.set_request_failed_async(request_id, e)
660
662
  logger.error(f'Request {request_id} failed due to '
661
663
  f'{common_utils.format_exception(e)}')
662
664
  return True
@@ -671,13 +673,13 @@ async def _execute_request_coroutine(request: api_requests.Request):
671
673
  except asyncio.CancelledError:
672
674
  # Current coroutine is cancelled due to client disconnect, set the
673
675
  # request status for consistency.
674
- api_requests.set_request_cancelled(request.request_id)
676
+ await api_requests.set_request_cancelled_async(request.request_id)
675
677
  pass
676
678
  # pylint: disable=broad-except
677
679
  except (Exception, KeyboardInterrupt, SystemExit) as e:
678
680
  # Handle any other error
679
681
  ctx.redirect_log(original_output)
680
- api_requests.set_request_failed(request.request_id, e)
682
+ await api_requests.set_request_failed_async(request.request_id, e)
681
683
  logger.error(f'Request {request.request_id} interrupted due to '
682
684
  f'unhandled exception: {common_utils.format_exception(e)}')
683
685
  raise
sky/server/requests/payloads.py CHANGED
@@ -363,9 +363,10 @@ class CancelBody(RequestBody):
363
363
  return kwargs
364
364
 
365
365
 
366
- class ClusterNameBody(RequestBody):
366
+ class ProvisionLogsBody(RequestBody):
367
367
  """Cluster node."""
368
368
  cluster_name: str
369
+ worker: Optional[int] = None
369
370
 
370
371
 
371
372
  class ClusterJobBody(RequestBody):
sky/server/requests/preconditions.py CHANGED
@@ -90,7 +90,7 @@ class Precondition(abc.ABC):
90
90
  while True:
91
91
  if self.timeout > 0 and time.time() - start_time > self.timeout:
92
92
  # Cancel the request on timeout.
93
- api_requests.set_request_failed(
93
+ await api_requests.set_request_failed_async(
94
94
  self.request_id,
95
95
  exceptions.RequestCancelled(
96
96
  f'Request {self.request_id} precondition wait timed '
@@ -98,13 +98,15 @@ class Precondition(abc.ABC):
98
98
  return False
99
99
 
100
100
  # Check if the request has been cancelled
101
- request = await api_requests.get_request_async(self.request_id)
101
+ request = await api_requests.get_request_async(self.request_id,
102
+ fields=['status'])
102
103
  if request is None:
103
104
  logger.error(f'Request {self.request_id} not found')
104
105
  return False
105
106
  if request.status == api_requests.RequestStatus.CANCELLED:
106
107
  logger.debug(f'Request {self.request_id} cancelled')
107
108
  return False
109
+ del request
108
110
 
109
111
  try:
110
112
  met, status_msg = await self.check()
@@ -116,7 +118,7 @@ class Precondition(abc.ABC):
116
118
  self.request_id, status_msg)
117
119
  last_status_msg = status_msg
118
120
  except (Exception, SystemExit, KeyboardInterrupt) as e: # pylint: disable=broad-except
119
- api_requests.set_request_failed(self.request_id, e)
121
+ await api_requests.set_request_failed_async(self.request_id, e)
120
122
  logger.info(f'Request {self.request_id} failed due to '
121
123
  f'{common_utils.format_exception(e)}')
122
124
  return False
@@ -166,7 +168,10 @@ class ClusterStartCompletePrecondition(Precondition):
166
168
  api_requests.RequestStatus.RUNNING
167
169
  ],
168
170
  include_request_names=['sky.launch', 'sky.start'],
169
- cluster_names=[self.cluster_name]))
171
+ cluster_names=[self.cluster_name],
172
+ # Only get the request ID to avoid fetching the whole request.
173
+ # We're only interested in the count, not the whole request.
174
+ fields=['request_id']))
170
175
  if len(requests) == 0:
171
176
  # No running or pending tasks, the start process is done.
172
177
  return True, None