I woke up today at a time I never thought I’d see again: 8am. As someone who usually wakes up around 5:30-6am unable to return to sleep, it was definitely a welcome surprise. But more intriguing was the sleep chart from my Fitbit.
Luna has acquired a new skill that I can only describe as “yowling in utter distress”. This has been happening for a while now, but she’s been steadily increasing the volume and urgency with every passing day. With that said, that sounds a lot worse than what’s actually happening.
So now that I have the GMMK, I decided to see just how bad my old Logitech G710 keyboard was after five years of service. It… wasn’t pretty. Imagine five years of never once cleaning it in a four cat household, and never even doing the usual cotton swabs soaked in rubbing alcohol every few months. Well, imagine no further!
I’ve considered this once or twice in the past, and I’ve come to the conclusion that Colleges and Universities may very well end as an institution. They’ve long since outlived their usefulness as centers of collaboration and education and devolved into mere adult daycares.
In the process of working on a customer migration and came up with a neat query for finding tables that don’t have a primary key:
SELECT c.oid::REGCLASS::TEXT AS table
FROM pg_class c
JOIN pg_namespace n ON (n.oid = c.relnamespace)
LEFT JOIN pg_constraint p ON (p.conrelid=c.oid AND p.contype = 'p')
WHERE c.relkind IN ('p', 'r')
AND n.nspname NOT IN ('pglogical', 'information_schema', 'bdr')
AND n.nspname NOT LIKE 'pg\_%'
AND p.conrelid IS NULL;
For anyone out there that is using pglogical, this is how you figure out if there are any tables that won’t work in the default replication set.