I've been running a cronjob for a while that runs repoman and collects the output. The per-category output of "repoman full" can be seen here: http://packages.gentooexperimental.org/repoman-checks/ Using grep I collect some of the more interesting errors: http://packages.gentooexperimental.org/repoman-current-issues.txt And then the new issues for each run get sent to #gentoo-bugs so that people can respond in a timely fashion. I've documented the rough outline here: http://gentooexperimental.org/~patrick/weblog/archives/2013-07.html#e2013-07-08T06_59_41.txt It'd be really nice to have that machinery run on infra hardware ... (It takes about 130 cpu-minutes at 3.4Ghz to run, but parallelizes perfectly)
@infra: I assume this can be closed as won't fix or obsolete since CI has been set up a while ago and I doubt infra wants to run a repoman version as well.
Even if, I would dare say it won't happen if it didn't happen in 6 years, and nowadays repoman is on life support only.