[galaxy-dev] DRMAA/SGE job handling regression?

Hello all, On our main Galaxy tracking galaxy-dist using DRMAA/SGE, jobs submitted to the cluster and queued and waiting (qw) are correctly shown in Galaxy as "grey" pending entries in the history. With my test instance tracking galaxy-central (along with a new visual look and new icons), such jobs are wrongly shown as "yellow" (running). Is this a general regression affecting other people? There also seem to be issues where killing a job in Galaxy just hides it but it remains running (yellow once you tick show deleted datasets, and running on SGE too). This was working properly on galaxy-dist (the job was killed on the cluster, and shown as red if you ticked show deleted datasets). Thanks, Peter