Hi, Andreas On Wed, 5 Jun 2024 18:39:40 +0200 Andreas Metzler <ametz...@bebt.de> wrote: > Package: curl > Version: 8.8.0-1 > Severity: serious > Tags: ftbfs > Justification: fails to build from source (but built successfully in the past) > > I did a local rebuild on sid which failed: > srcdir=. /usr/bin/perl -I. ./runtests.pl -a -p ~flaky ~timing-dependent -n > -j35 ~300 ~301 ~303 ~304 ~306 ~309 ~310 ~325 ~364 ~400 ~401 ~403 ~406 ~407 > ~408 ~409 ~410 ~414 ~417 ~560 ~678 ~987 ~988 ~989 ~1112 ~1272 ~1470 ~1561 > ~1562 ~1630 ~1631 ~1632 ~2034 ~2037 ~2041 ~3000 ~3001 ~3102 > Using curl: ../src/curl > [...] > TESTDONE: 1700 tests were considered during 522 seconds. > TESTINFO: 195 tests were skipped due to these restraints: > TESTINFO: "curl lacks debug support" 99 times (159, 356, 358, 359, 363, 412, > 413, 437, 438 and 90 more) > TESTINFO: "curl lacks unittest support" 49 times (1300, 1302, 1303, 1304, > 1305, 1306, 1308, 1309, 1323 and 40 more) > TESTINFO: "configured as DISABLED" 16 times (323, 594, 836, 882, 938, 1182, > 1184, 1209, 1211 and 7 more) > TESTINFO: "failed starting HTTP/2 server" 7 times (1700, 1701, 1702, 2402, > 2403, 2404, 2405) > TESTINFO: "curl lacks Schannel support" 6 times (2033, 2070, 2079, 2087, > 3023, 3024) > TESTINFO: "no gnutls-serv (with SRP support)" 4 times (320, 321, 322, 324) > TESTINFO: "curl lacks http/3 support" 3 times (2500, 2502, 2503) > TESTINFO: "curl has idn support" 3 times (959, 960, 961) > TESTINFO: "curl lacks TrackMemory support" 2 times (96, 558) > TESTINFO: "curl has ipv6 support" 1 time (1454) > TESTINFO: "curl has threaded-resolver support" 1 time (506) > TESTINFO: "curl has proxy support" 1 time (375) > TESTINFO: "openssl engine not supported" 1 time (307) > TESTINFO: "curl lacks NTLM_WB support" 1 time (1310) > TESTINFO: "curl lacks Unicode support" 1 time (1703) > IGNORED: failed tests: 587 > TESTDONE: 1502 tests out of 1505 reported OK: 99% > > TESTFAIL: These test cases failed: 241 507
Were you able to consistently get those exact failures? We are getting some random and unreproducible failures since we have enabled parallel tests - but the speed up is jaw dropping. We have reported it to upstream already [1], but it's not very easy to debug. I also tried to build locally with sbuild and using debci [2] and they have succeeded, so maybe it's just a hicup. If you were able to consistently reproduce this, please let us know more about your build environment. Cheers, Charles [1] https://github.com/curl/curl/issues/13350#issuecomment-2119091806 [2] https://ci.debian.net/data/autopkgtest/unstable/amd64/c/curl/47382526/log.gz
signature.asc
Description: PGP signature
