Red Hat Bugzilla – Bug 1216096
Hard coded /tmp size of 64M causes problems
Last modified: 2015-05-25 08:01:49 EDT
+++ This bug was initially created as a clone of Bug #1215768 +++
I'm looking at b482aff8fbb5dc59d25335b67353465071d6bd45 in rhatdan/docker
It seems that you are hode coding all tmpfs mounts to 64M. Docker previously mounted tmpfs at /tmp and this added /run.
Golang uses TMPDIR to do its builds. When doing a rather large build (like openshift) which is larger than 64M it obviously can not work!
I think this size needs to be somehow configurable... Deciding that X is enough /tmp space for everyone just doesn't seem like a possibility...
--- Additional comment from Daniel Walsh on 2015-04-28 08:13:12 EDT ---
Eric, any chance of changing TMPDIR -> /var/tmp?
--- Additional comment from Daniel Walsh on 2015-04-28 08:15:46 EDT ---
This should also not be happening in a docker build?
--- Additional comment from Daniel Walsh on 2015-04-28 08:32:57 EDT ---
Removed the size restrictions on /tmp and /run
Lokesh we need a new build of docker-1.6
docker-io-1.6.0-4.git350a636.fc21 has been submitted as an update for Fedora 21.
* should fix your issue,
* was pushed to the Fedora 21 testing repository,
* should be available at your local mirror within two days.
Update it with:
# su -c 'yum update --enablerepo=updates-testing docker-io-1.6.0-4.git350a636.fc21'
as soon as you are able to.
Please go to the following url:
then log in and leave karma (feedback).
docker-io-1.6.0-4.git350a636.fc21 has been pushed to the Fedora 21 stable repository. If problems still persist, please make note of it in this bug report.
I wonder if this fix is what is causing https://bugzilla.redhat.com/show_bug.cgi?id=1213258