Widower

Just a PSA: tonight we did some emergency maintenance that hopefully should keep the issue from crashing the servers again. In commemoration, double XP is available on all shards until at least Tuesday. We are considering what to do for those who lost anything on account of the crashes.

So, over the past 24 hours we've had a few crashes take down Excelsior and Everlasting (you know, the servers most of you play on). Just in case anyone was wondering, this is unrelated to the issue from the weekend (or clicking too many doors in Trick or Treat :P). The issue is that vRack (the LAN which lets the servers in the cluster talk to each other) keeps failing, for reasons unknown. Hopefully OVH will clear up the issues sooner rather than later, but until then we're kind of at their mercy.

Everlasting and Victory are up on one of the other hosts. We'll keep working to bring the affected host back up, though the ball is largely in OVH's hands. Hopefully it's back up for Tuesday maintenance and we'll rearrange everything then. Shout if you notice anything awry. Otherwise, go, hunt, kill skuls.

Okay, just to recap for those whose days haven't been consumed by this: VM3 went down around 3:15 Eastern. We spent a couple hours trying to see if we could prompt OVH to do something about it, which eventually succeeded. They replaced a CPU in the machine, but when Cipher went to reconnect it, IPMI wasn't working, so we ended up spending another three hours waiting for OVH to do something. It took them about two hours to replace the motherboard, RAM and CPU cooling system, but when Cipher went to reconnect it (yes, something of a pattern here) he found that some virtual disks were missing. Nothing mission critical, but still not good. Until we can figure out what's going on there, VM3 remains out of commission. Cipher set up VMs for Everlasting and Victory on one of the other hosts during one of the waits, so he's spinning those back up and is hoping to have them back up that way. If you were playing at the time of the initial crash, you shouldn't experience more than 17 minutes of rollback, and hopefully much less.

From Cipher: Small (non) update - unfortunately the server appears to be booted into OVH's DC tech image still and the remote access module for the server is completely unresponsive so we're unable to reboot it or change any configuration on the host. OVH automatically identified the issue and sent us a notice that they'd be intervening within a few minutes, but that was over two hours ago and we haven't heard back since. This will be the last update for tonight unless the server becomes accessible again. Thank you for your patience. In other words: we're still waiting on OVH. We don't anticipate it getting fixed tonight at this point, so go nuts with the Yins. We will give ample notice if something changes and we are able to restore the other servers.

We hit a snag with the reboot and OVH is looking at it again. Once that's done, we will hopefully bring the host back online. You can keep playing on the functioning shards, but we wouldn't recommend starting a TF or anything at the moment, since there's a decent chance everyone gets disconnected when we bring up the affected host.

We've confirmed that OVH is actually working on the issue now!

From Cipher: We've been informed by OVH support that we're now unable to request intervention for any incidents including those resulting from hardware failures outside of OVH's standard technical support hours without paying for Business support at a cost of 10% of our monthly bill ($485/mo). Unfortunately, this means we're unable to request intervention until Monday morning. That said, it appears that the host was flagged for intervention either by a technician or by the support agent we spoke to, but neither has been communicated to us at this time. We're evaluating our options while we wait out the result (or lack thereof) of the intervention.

Hello! As you may have noticed, everything went down about half an hour ago. This is because one of our physical hosts has gone out for reasons currently unknown, and unfortunately that one had the auth server on it. In the meantime, we have migrated the auth server to a different host so that you can keep playing. The one snag is that Everlasting and Victory are on the affected host, so they will be down until it's fixed. We hope to restore service to those servers Soon™. Everything else should be up - ignore the server status if it says everything is down, it's reading from the bad host. Thanks!

Servers other than Everlasting and Victory should be accessible shortly.

We are aware something is up and we are working on it!

Hello! It's October and Atlas is suddenly holding up a big pumpkin, so you must be asking, "When is Trick or Treat?!?!" I am pleased to announce that the Halloween event is being activated today, October 1st, and will run until November 4th. Your next question: "Did anything change this year?" No changes have been made to the event for this year; everything from last year's event post is still correct. The devs have been busy with other matters (not the pirates, different other matters) which we hope to be getting to you Very Soon™. What could be more terrifying than that? In the interest of transparency, we would also like everyone to be aware we have lowered the player limit on Peregrine Island to 50 to make Trick or Treating a bit easier on everyone. We will keep an eye on this as it develops and see if it needs further adjustments. Thanks for reading! Enjoy the event!

No. (Which is also a pretty good answer to the thread title, if you'd like more to debate.)

Also the Focused Feedback threads, the Temnix and Diantane threads, anything about Sentinels... we could fit a lot of stuff in this category, really.

I've been thinking since this series of threads got started I should really just rename the place to "Forum PVP," but that works too.