Comment 13 for bug 1559723

Some relevant lines from /var/log/syslog:

Mar 19 19:15:10 server smartd[2251]: Device: /dev/sdb [SAT], SMART Usage Attribute: 194 Temperature_Celsius changed from 107 to 106

Mar 19 07:59:21 server console-kit-daemon[2041]: GLib-CRITICAL: Source ID 376 was not found when attempting to remove it
Mar 19 19:23:17 server console-kit-daemon[2041]: GLib-CRITICAL: Source ID 332 was not found when attempting to remove it
Mar 19 19:23:17 server console-kit-daemon[2041]: GLib-CRITICAL: Source ID 332 was not found when attempting to remove it
Mar 19 19:23:41 server console-kit-daemon[2041]: WARNING: Record was not written to disk (No space left on device)

Mar 19 19:24:39 server console-kit-daemon[2041]: message repeated 3 times: [ WARNING: Record was not written to disk (No space left on device)]
Mar 19 19:45:07 server console-kit-daemon[2041]: GLib-CRITICAL: Source ID 384 was not found when attempting to remove it
Mar 19 19:45:11 server smartd[2251]: Device: /dev/sdb [SAT], SMART Usage Attribute: 194 Temperature_Celsius changed from 106 to 107

So besides full disk and maybe broken file system, the temperature of one drive also was too high, presumably during the balance operation. After >10 hours of idling, temperature is normal again (46 degrees celsius):

smartctl -a /dev/sdb
194 Temperature_Celsius 0x0022 104 101 000 Old_age Always - 46