Last modified: 2014-09-17 22:12:39 UTC
After updating and labs-vagrant provision of a flow-tests labs instance, I was getting 503 errors. /var/log/hhvm/error.log contained [Wed Sep 17 20:41:47 2014] [hphp] [24230:7fb9ed1fe6c0:0:000001] [] Failed to initialize central HHBC repository:\n Failed to initialize schema in /var/run/hhvm/hhvm.hhbc: RepoQuery::step(repo=0x7fb9dd467000) error: 'CREATE TABLE main.writable_4cd91499a6638cdf8f21e85093dc7dab65e909e0(canary INTEGER);' --> (13) database or disk is full\n Failed to open /var/www/.hhvm.hhbc: 14 - unable to open database file\n Turns out /var/run/hhvm/hhvm.hhbc was 200MB which filled the tmpfs on /run that is only 201MB on this small labs instance. The workaround is to delete hhvm.hhbc. Note that you need to stop and start both hhvm and jobrunner, because both services have a handle to this file. There was IRC discussion of placing hhvm.hhbc in a better location, e.g. /srv somewhere.
