On 2013-02-21 12:12, Holger Levsen wrote: >> It probably causes an issue with open file descriptors: only plots for >> the first 62 sections are generated, thereafter rpy fails: > > we dont have that many sections do we?
I do. And I'm missing one quarter of the plots :-( > (And even if we had, I rather consider this bug "normal", since it doesnt > have > a major impact on the usability of piuparts for everyone.) It affects an 'important' part of all existing piuparts master/slave setups, about 1/4. (Are there more instances than piatti, goldwasser, Dave's and mine?) And having the rpy jobs only finish at the end is probably also a cause for the vast memory usage of piuparts-report, at the end it currently takes nearly 6 GB ... Andreas -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
