On Fri, Jun 19, 2020 at 04:52:30PM +0100, Michael wrote: > On Friday, 19 June 2020 16:35:27 BST urp...@gmx.com wrote: > > On Fri, Jun 19, 2020 at 09:45:59AM -0500, Dale wrote: > > > Jack wrote: > > > > Not far enough back. You need to show the actual error, not the line > > > > "Error1:" which gets printed after the error. > > > > > > And depending on the number of cores/threads and other emerge settings, > > > it can be a dozen, two dozen lines or sometimes even further back than > > > that. With CPUs having a dozen or so cores/threads, it's amazing that > > > some stuff compiles at all. Prime example, recent thread about Pam > > > updates. > > > > > > OP, if needed, tar the whole error log and attach it. Just don't post > > > it elsewhere and link to it tho. After a while, it's gone or a person > > > finds the log but not this thread with the solution. > > > > > > Dale > > > > > > :-) :-) > > > > Ah ha. Yes Thank you. Sorry, I've never been good with e-mail:-P > > > NP, did you get an OOM error in dmesg when this happened by any chance? You > don't seem to have much RAM, you have no swap and the number of jobs is > relatively high for memory hungry compiles. > > It could be a bug, but unless a report has been filed already in BGO to this > effect, I suggest you add a swapfile, enable it, and try again with > MAKEOPTS="-j1" or "-j2". Please ask if you need more detail.
I was thinking about -j1. I tried -j4. I thought I didn't need swap with >=4GB of RAM, but it makes sense. My machine kinda crawls trying to build that package. Oh man, I'll have to re-partition:-(
