migration to new server

[abma]

most services are migrated to the new server already, i've migrated mysql today. hopefully this causes no problems as some config settings has to be changed and mysql is running on linux (again).

things left is the windows vm which hosts zero-k.info / springie, which hopefully can be moved soon to the new server, too.

sorry for the downtimes!

[PicassoCT]

Always them serverimmigrants stealing our games,woman, jobs and education!

[dansan]

Thank you for all that work!

Can you announce a downtime (even if it is a small one) of the lobby server next time shortly before in #main?

[Licho]

I have announced downtimes with broadcast.. what lobby did you use?

[abma]

are announcements only sent to online players?
if so, MOTD have to be set, too?!

[abma]

we've lost some lobby accounts again, i'll try to migrate tomorrow to the mysql backend with recovering the lost accounts.

maybe see https://github.com/spring/uberserver/issues/13 for more info.

[dansan]

I have announced downtimes with broadcast.. what lobby did you use?

SpringLobby.
"broadcast" means I should have gotten a PM?

Maybe I just didn't see it then... can you tell me (parts of) the text in the broadcast, so I can grep in my chatlogs. Then I'll be able to answer that question - if it is of interest.

migrate to mysql backend

Nice :)

[Silentwings]

Thanks for doing all this work!

On Springlobby, I didn't see any announcements of downtime (except the messages the server sends 1 sec before it restarts).

[Licho]

Thats it silentwing, its unknown when/if restarts will be needed.

[Silentwings]

Aha - no problem :)

[Jools]

Thanks for doing all this work!

[dansan]

Then let me rephrase my request:
I think it'd be better for the users (players) to be notified some minutes or hours before a restart. You'd do that on a professional server, if a downtime cannot be avoided, too.

Can the broadcast be made manually by an admin/moderator? Then abma/Licho could do it some time before a restart, so people are a little prepared.

Additionally a little red warning notice on the top of springrts.com/ and zero-k.info/ about a planned server downtime would be good too.

Please don't understand me wrong: I'm not picking on you admins, but I think it's important for better communication with your users!
They should not think "uh - server broke down cause its poor software or bad dev/admin", but they should think "admin works for me and upgraded server so I can keep having fun".

There is a big difference for the user between "random restart" and "working on a better experience". Let them appreciate your work!

[Jools]

Yes, but I did get broadcast messages "server restarting". Maybe everybody didn't though.

[knorke]

Made a global notifaction thread. Bit late but the first time I tried forum went down when I wanted to post.

The only lobby broadcast I got in springlobby was the test one. (in red)
[23:32] <[2up]knorke> next time a notice maybe
[23:32] <Licho> i made adminbroadcast
[23:32] <Licho> was it not visible?
[23:33] ** Server ** s his visible?
[23:33] <[2up]knorke> this is visible
[23:33] <Licho> safe mode is in recovery mode?
[23:33] <[2up]knorke> before was just "connection lost"
[23:33] <Licho> ok so its broadcast and not adminbroadcast

[abma]

i've "migrated" the lobbyserver to the mysql backend, hopefully now everything works as before.

i'm still expecting some weird bugs, as it wasn't tested under load, so please report newly introduced bugs to https://github.com/spring/uberserver/issues

thanks!

some springie hosts seems to be confused, they try to login twice, but they are already logged in and respond to !help and stuff...

[dansan]

Made a global notifaction thread.

Thanx! Can you make the news appear on http://springrts.com/ ?

[Floris]

May i notify that the irc bridge is down for a couple of days now?

[abma]

You sure? This works: http://webchat.springrts.com/, so it should be up.

[abma]

these accounts where lost on 20130919 on the old server, i've imported it from the backup of the old server:

GargantuaSauce, wiremgon, [S44]Nemo, [Evo]ForbodingAngel, github, nixtux_lostpassword, AlcariTheMad, Floris

i had to recover / import from multiple backup files of the old server, hopefully i've imported now all users into the new database. if not, please report!

looks a bit like we are at the point of no return

[Anarchid]

github

that nicely explains something that puzzled me for a while:

[16:34:25]
github
[spring] none pushed 3 new commits to develop: http://git.io/Sz38Hw
[16:34:25]
spring/develop 44c51c3 jK: fix a severe LuaHandleSynced bug
[16:34:25]
spring/develop 0d01c45 rt: assimp: fix major issue with baked transforms
[16:34:25]
spring/develop 033c93c rt: ws