So, today I have had the following problems on various databases:
1) lack of hosts file causing intermittent connectivity
2) running out of space in innodb tablespace
3) a backup causing undue slowness
4) a data partition running out of space and corrupting the binary log, thus making the slaves not replicate
5) mysql server not starting after I added space (see 2) because I put the same logfile in twice
All in all, not bad. Only problem 4 caused me to furrow my brow to fix it, because I hadn’t seen anything like it on the list, but one of the first google hits led me to the cause. Basically the symptom was a slave server that would not start replication, with the error in mysqld.err only (not in SHOW SLAVE STATUS — that would always show NULL seconds behind). But Google immediately gave me the answer.
What a day. Luckily I slept for 11 hours last night; I’ve been working 12 so far!