FANNG1 commented on PR #67553: URL: https://github.com/apache/doris/pull/67553#issuecomment-5630855853
The current CI failures (`FE UT` and `check_coverage_fe`) look unrelated to this PR. Both are caused by the build environment image, which now ships glibc 2.17. **Image change** `apache/doris:build-env-ldb-toolchain-latest` was re-pushed on Docker Hub at 2026-09-10 16:10 UTC: | Run | Build | Image digest | Result | |---|---|---|---| | FE UT, before | [1041974](http://43.132.222.7:8111/viewLog.html?buildId=1041974&buildTypeId=Doris_Doris_FeUt) | `sha256:84a9b554fecd8dc2998879a71f2d641e2da8dab45dc5b0cef93f2707b5776e01` | passed (9160 tests) | | FE UT, after | [1043001](http://43.132.222.7:8111/viewLog.html?buildId=1043001&buildTypeId=Doris_Doris_FeUt) | `sha256:4b35f2346ceeb072ae842bfbd3484d52ecc63a1ee19953a194b062dae775f893` | 17 failures | Both runs tested the same commit `829c628de9e177fc44bff1f5830bcf8f87413c5c`. **FE UT (build 1043001)** All 17 non-muted failures are in `LanceIndexMetadataLoaderTest` (13) and `LanceRestCatalogTest` (4). Neither class is touched by this PR. The Lance JNI library fails to load: ``` java.lang.UnsatisfiedLinkError: /tmp/liblance_jni11342793688463938227.so: /lib64/libm.so.6: version `GLIBC_2.27' not found (required by /tmp/liblance_jni11342793688463938227.so) ``` After that, the remaining Lance tests fail with `java.lang.NoClassDefFoundError: Could not initialize class org.lance.Dataset` / `org.lance.namespace.RestNamespace`. `LanceScanNodeTest`, the test changed by this PR, passes 21/21 in the same run. **check_coverage_fe (build 1042299)** The container prints `glibc 2.17`, and then Node 24.19.0 installed via nvm cannot start, so the script exits before `build.sh --fe` runs: ``` node: /lib64/libc.so.6: version `GLIBC_2.28' not found (required by node) node: /lib64/libstdc++.so.6: version `GLIBCXX_3.4.21' not found (required by node) ``` Other PRs are hitting the same `check_coverage_fe` failure right now (for example #67648, #67820, #67627, #67784, #67810). Could someone with CI access take a look? Rolling the tag back to the `84a9b5...` image, or pinning the CI to an image digest, should fix this. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
