So I don't have access to any of the Travis slaves - that's all managed by Travis-CI the company. I do have Admin access to the github org and by extension to all of the travis settings, but not employed by travis.
On Thu, Jun 25, 2015 at 5:40 PM, Rafael Fonseca <rsafons...@gmail.com> wrote: > Hi David, > How much permissions do you have on the Travis Infra? > It seems some slaves have a full /tmp and are causing some failures due to > that... can you fix it, or should i just draft a command that will try to > safely cleanup the /tmp files before build starts ... something like rm -f > all files older than x days and that are not used by any process based on > lsof output? > You can take a look at hard evidence of this problem here -> > https://travis-ci.org/apache/cloudstack/jobs/68291716 > Expand before_script.sh and scroll down to line 280: > src/lxml/lxml.objectify.c:35413:1: fatal error: error writing to > /tmp/ccSvJbDa.s: No space left on device > > > > On Thu, Jun 25, 2015 at 11:12 PM, David Nalley <da...@gnsa.us> wrote: > >> Today, no. >> Supposedly the GH folks are going to add a PR/GH only permissions >> level, but that isn't yet available. >> >> If you have a list of ones you'd like me to kick off, I can do so. >> >> >> On Thu, Jun 25, 2015 at 8:17 AM, Daan Hoogland <daan.hoogl...@gmail.com> >> wrote: >> > Anybody knows if I can restart travis builds that failed in the past >> > but should pass IMNSHO? >> > >> > I want to trigger some to see if I can merge them. I am not the owner >> > of the pull request nor, obviously, of the ACS repo. >> > >> > -- >> > Daan >>
