This is expected if you try to update the repository at the same time the daemons are updating it. Is it possible that's the issue?

Well the daemons were failing with the same message. I tried running it manually to see if phd was at fault at all. Same stacktrace either way.

You can figure out what's holding the lock by running this command:

mysql> SELECT IS_USED_LOCK('global:phabricator:PhabricatorRepositoryManagementUpdateWorkflow:SVN');

If the lock is currently held, that will return a connection ID. You can find more information about the connection in:

mysql> SHOW FULL PROCESSLIST;

The connection IDs are listed in the first column. The "Host" might be enough to tell you what's holding the lock, or you could use something like netstat -t to figure out which process is connected on the given port.

In general, this behavior is expected if multiple processes are trying to update a repository. It just means "some other process is updating this repository right now, and holding a global lock in MySQL while it does". This can happen if you accidentally launch several PullLocal daemons, or someone runs a rogue PullLocal daemon somewhere, or multiple users are running repository update, or some combination of those things.

Cool. I'll repurpose this to make the exception more clear.