Darn it - I was wondering about the best place to post feature requests.

Logs from a vanilla build, while working with @Chandler. Only changes were copying our before-deletion CATALOG.cat into world0/ and the complete respective blueprints/ structure as well.

Player counts has not affected lag @Chandler.

Complete log.txt.3 file, additional logs available upon request (will retain for at least 30 days)

I'm the owner of stars.bitminery.com:4242 We can also replicate this in our staging environment with a vanilla build (+ a copy of production's blueprint DB).

Charge occurs at time of blueprint purchase, just tested this.

There is a cost for station blueprints, using normal settings. Although ship blueprints are free.

I have mixed feelings about the position taken here.

I should probably point out that this NullPointer was initially discovered while investigating odd server slowdowns. The error will often be written dozens of times per second to logs, and that has the apparent effect of impairing response times for all players.

This is an exported sector of where you'll see the NullPointerException in server logs, upon shooting the station within.

Added @spunkie as we were working together to flush-out an addition to T1385 at the time.

Editing...it may have been premature for me to assume this bug is the same. I've created a new one: http://phab.starma.de/T1388

@spunkie invested some time chasing this bug with me last night. Tagging him for inclusion.

Here's a little more info on what pilots were doing at the time the server was most recently forcefully shutdown due to a detected deadlock: we have automated 5 minute checks (via analysis of jstack dumps). If 1st check detects a deadlock we attempt an orderly server shutdown, failing that the server's java instance is forcefully killed after 2 additional minutes. The intent of a forceful shutdown is mostly to minimize pilot impact: deadlocks seem to prevent any proceeding saving of work, until the server is restarted.

Just wanted to followup. We've seen a number of additional deadlock conditions since reporting this bug. One is tagged as a duplicate to this task. I haven't submitted the additional ones. These are very disruptive to our player-base, especially since the crashes result in world rollbacks of around 15-20 minutes; lost ships, lost materials, and pilots respawning to find themselves spaced, and their ship multiple sectors away.