The cluster shutdown has been temporarily cancelled.
It’s not even the day of no DT but the server already has aneurysms 2 days in a row.
1 Like
“Cluster Shutdown Alert: Cluster Shutdown Delayed” at 13:09. Is this notification about the shutdown tomorrow being pushed back? There is no threat of a second downtime coming soon?
so first of all … its normal that the graph shows all the time that around 10:40 UTC drops the activity on TQ … everyone docks on stations or goes offline because the downtime is close. writing something like “yea its always around this time mostly inactive” is not really smart. cause players are used to the downtime causing automaticly a drop everyday at that time. if you would switch the down time one hour back or forth and keeping that for 1 year and the player base gets used to it would the same phenomenom occur that for example now at 11:40 drops suddenly the activity and the usual down time right now would be perfect …
in short writing down “everyone gets inactive already to the current downtime” was a stupid thing to do, reason is written down.
so, no belt respawn then? so the npcs will clear all the belts? i recommend to remove or lower the frequency of npc mining ships. with that staying the belts maybe longer alive.
or why not just do auto respawns like anomalies? that would solve the trouble with the belt completely.
iunno about the marketplace tho =w= i never thought no downtime would actually affect the eve market =x i thought it always updates like every hour or something like this x…x
2 Likes
Asteroid Belts/Anomalies/Wormholes etc you can compare to instances. Forget the realms as you are nitpicking. I mentioned it only as an idea. Don’t forget if EVE would have such a large player base as WoW they also would have to split their servers into what you can compare to realms to handle the player load. A realm is a particular server cluster for your info.
1 Like
Aussies and kiwis are not generally known to be quiet complainers, y’all simply have a unique vocabulary for describing discomfort 
1 Like
It also used to happen randomly a lot more because live fixes were not a thing in games in those days. Compared to then, it’s heaven now.
I think CCP will be in for a surprise at how many artifacts are left with the game client when it disconnects. Using down-times to flush them is poor coding so I hope a virtual laundry list of poorly documented systems will rear their ugly heads; we can all reply with “See? Have we not been complaining about this =for years=?”
4 Likes
If no actions are taken by the player’s client for x minutes that’s considered afk and the connection is severed. This has been a MMO standard for like 15 years. Eve is honestly the only game I can think of that doesn’t have afk timeout except specific games that use afk players as bazaars.
ikr. I’m for this driving it til the wheels fall off. Cuz the wheels are definitely falling off this tricycle.
This Test will answer that .
is it, from a technical and practical standpoint, possible to have no downtime at all?
I wanna say yes, but currently eve is like:
Yes and no. Yes, it is entirely possible to inject live fixes that have to do with player data. No, it’s not possible to touch a game system without a hard reboot.
Having a one time, one day experiment to gather data which will be used to determine issues and priorities for devs to focus their effort on seems to be a good choice, especially with a more resource limited company. That being said, eliminating a DT of around 5 minutes shouldn’t be the goal; EVE has far too many previously documented deficiencies that need to be resolved at a much higher priority in the queue. For those concerned about a 5 minute time out to their gameplay, perhaps a regular,rolling DT that corresponds with the London technicians/EVE devs work schedule could spread the pain around.
1 Like
I think it has them, but I also think they’re broken 
Now cloaky campers, afk citidal gunners and other passive players effecting the game won’t even need to log in 
1 Like
i didn’t say it was down to PA, i asked. bit of a difference.
They have been saying they will do lots and lots of things for years now and they haven’t. But suddenly and mysteriously they seem to have gotten around to finally doing (a lot of terrible, truly terrible) things and all of it “is nothing but coincidentally” just after PA get their fingers in the pie?
Really?
Do grow up.
1 Like
Change your brand of tin foil. PA is not involved in the day-to-day operations of EVE.
They’ve been steadily making changes on a variety of things. Ask anybody who has been gone for a year what it’s like when they come back into the game and how they have to get back up to speed.
1 Like
First, let’s take a step back and look at the reasons why we have downtime in the first place:
- We have excessive memory consumption and lack of clean-up in certain areas, and we don’t necessarily refresh cache since the daily reboot will take care of it all.
- We still have daily database jobs that run during downtime.
- There are certain things that must be done regularly, and it is most convenient to do them during startup when there are no players online.
This is our dirty laundry. We have documented all the things we know that can go wrong. Then there are the things we don’t know about and testing is the only way to find out…
For some of the people on here skeptical of whether or not this will deliver value… Technical debt is a thing that can hold back an entire system, and any codebase as old as EVE is going to have a lot of TD.
Sounds like there’s a lot of TD built up around, or reliant on, downtime as a failsafe.
Even if “removing downtime” isn’t the end goal - it’s not really about that. It’s about finding the crappy parts of the codebase that can and should be better.
It’s boring meat-and-potatoes stuff, but it can be astonishing the improvements you get to seemingly unrelated parts of a codebase when you polish up and improve the engineering of basic systems…
And memory management is a basic system. I winced when I saw that they aren’t bothering with memory management a lot of the time because of downtime. I firmly hold the posiiton that, in software design, there’s no such thing as “best practice” - there’s only contextually appropriate tradeoffs. But if there was such a thing as “best practice”, ignoring memory management because a nightly reset will fix it wouldn’t be it.
Owning up to the dirty laundry and taking steps to test and improve it are good things.
Additionally, dealing with in-game downtime-dependent gameplay would be a great boost too. For example, belt respawns being dependent on downtime gives big advantage to miners who can be online just after downtime. Spreading these things out to be random or semi-random would be beneficial.
At the same time, I have quite a lot of ore in Jita on three month sell prices waiting for price spikes. Crossing my fingers that I’ll get to clear some of it out after the belts are empty… But of course, market PvPers know it’s coming too. Will be interesting to see if I get a win out of it.
4 Likes
Could this be to identify Bots? Since whomever set them up would know the times, they would probably schedule breaks in runs around the DT. See if BotA docks and logs off before the downtime…
Though one day test wouldn’t tell that much…