Source: r-cran-bbmle Version: 1.0.25.1-1 User: debian...@lists.debian.org Usertags: regression
Hi Maintainer Sometime around 2024-04-18, r-cran-bbmle's autopkgtest regressed in testing on ppc64el [1]. I've copied what I hope is the relevant part of a recent log below. This appears only to be a warning, it would be good if this could be filtered out somehow. Regards Graham [1] https://ci.debian.net/packages/r/r-cran-bbmle/testing/ppc64el/ 72s --- profbound.Rout.save_ 2024-05-23 12:58:07.420821085 +0000 72s +++ profbound.Rout_ 2024-05-23 12:58:07.424820960 +0000 72s @@ -40,6 +40,9 @@ 72s + method="L-BFGS-B") 72s > 72s > pp <- profile(fit2,prof.lower=-0.2) 72s +Warning message: 72s +In mle2(minuslogl = function (a = NULL, b = NULL) : 72s + convergence failure: code=52 (ERROR: ABNORMAL_TERMINATION_IN_LNSRCH) 72s > stopifnot(min(subset(as.data.frame(pp),param=="b")$par.vals.b)==-0.2) 72s > ## note that b does go below -0.2 when profiling a ... 72s > options(old_opt) 72s autopkgtest [12:58:07]: test run-unit-test: -----------------------] 72s run-unit-test FAIL non-zero exit status 1