2.0 beta8: deployment with sphere as provider - bootstrap node - no space left on /

[Larry Michel]

While trying to collect juju log for bug 1592811, I am seeing / is out of space:

root@ubuntuguest:/var/log/juju# df
Filesystem 1K-blocks Used Available Use% Mounted on
udev 995416 12 995404 1% /dev
tmpfs 200076 488 199588 1% /run
/dev/sda1 10287416 10271024 0 100% /
none 4 0 4 0% /sys/fs/cgroup
none 5120 0 5120 0% /run/lock
none 1000364 0 1000364 0% /run/shm
none 102400 0 102400 0% /run/user

This could because I've left the system in this failure state for past 4-5 days, but the 10GB disk size seems rather small nonetheless for any of these deployed VMs.

root@ubuntuguest:/var/log/juju# ls -l
total 1619412
-rw------- 1 syslog syslog 314572625 Jun 21 01:28 logsink-2016-06-21T01-28-31.233.log
-rw------- 1 syslog syslog 314572710 Jun 21 02:31 logsink-2016-06-21T02-31-03.255.log
-rw------- 1 syslog syslog 271341172 Jun 21 04:10 logsink.log
-rw------- 1 syslog syslog 314572548 Jun 21 01:32 machine-0-2016-06-21T01-32-14.378.log
-rw------- 1 syslog syslog 314572619 Jun 21 03:00 machine-0-2016-06-21T03-00-19.607.log
-rw------- 1 syslog syslog 128618496 Jun 21 15:23 machine-0.log

[Cheryl Jennings (cherylj)]

Larry - were you able to change the disk size with constraints?

I also saw that the logs were taking up a large amount of space as you were running with TRACE.

[Cheryl Jennings (cherylj)]

Confirmed with Larry that he was able to specify a larger disk size using constraints.