Intermud-3 router old status

[cratylus]

Apparently the school campus suffered a power outage that hit most
buildings, including the one that *yatmim hides in. By the time I arrived,
power had been restored and the router was booting.

If you experience further problems, please let me know.

-Crat

[Atomic]

On my mud (in the Archroom) the Intermud router status says to be online,
but de printout says: Intermud3 link down, stats unavailable.
down, south (server monitoring room) says I3 Server monitoring (as well as OOB) is online, rest offline.

9 out of 10 that it is caused on my side of the connection... can someone confirm that I'm able to
receive global intermud conversations?

Code: [Select]

INTERMUD_D: prev: ({ OBJ(/secure/sefun/sefun), OBJ(/secure/daemon/ping) })
INTERMUD_D reloaded.
Loading object stack:
    0:"get_stack".OBJ(/secure/sefun/sefun)."get_stack"
    1:"create".OBJ(/daemon/intermud)."create"
    2:"CATCH".OBJ(/secure/sefun/sefun)."CATCH"
    3:"update".OBJ(/secure/sefun/sefun)."update"
    4:"CheckOK".OBJ(/secure/daemon/ping)."CheckOK"
    5:"heart_beat".OBJ(/secure/daemon/ping)."heart_beat"
Loading object trail: ({ OBJ(/secure/sefun/sefun), OBJ(/secure/daemon/ping) })
INTERMUD_D: SocketStat: 1
INTERMUD_D setup: ({ "startup-req-3", 5, "Netheria", 0, "*yatmim", 0, 0, -1, -1, 3000, 3005, 3008, "Dead Souls 2.4.2", "Dead Souls 2.4.2", "MudOS
    v22.2b14-dsouls2", "LPMud", "mudlib development", "mujo@net.nl", ([ "rcp" : 2990, "who" : 1, "emoteto" : 1, "http" : 2995, "channel" : 1, "finger" :
    1, "tell" : 1, "auth" : 1, "mail" : 1, "ftp" : 2999, "locate" : 1, "oob" : 3005 ]), ([ ]) })
ERROR RECEIVED: ({ "error", 5, "*yatmim", 0, "Netheria", 0, "not-allowed", "wrong password, and from a new IP", ({ "startup-req-3", 5, "Netheria", 0,
    "*yatmim", 0, 0, -1, -1, 3000, 3005, 3008, "Dead Souls 2.4.2", "Dead Souls 2.4.2", "MudOS v22.2b14-dsouls2", "LPMud", "mudlib development",
    "mujo@net.nl", ([ "who" : 1, "rcp" : 2990, "emoteto" : 1, "http" : 2995, "channel" : 1, "finger" : 1, "tell" : 1, "auth" : 1, "ftp" : 2999, "mail" :
    1, "locate" : 1, "oob" : 3005 ]), ([ "next boot" : "EST Thu Oct 18 00:48:02 2018", "oob port" : 3005, "upsince" : "Tue May 22 09:48:02 2007", "ip" :
    "127.0.0.1", "native version" : "2.4.2", "os build" : "unix" ]) }) })
errorcode: not-allowed

Hmm, not allowed on Yatmim anymore? "Wrong password and from another IP",
well that could be right, I have a DHCP ip so that may vary from time to time. Don't know about
any Yatmim-password though  (edit: checked faq's, gonna try a few other things http://lpmuds.net/intermud.html)

Switching to i4 as router, does seem to accept things, but still not receiving anything from the outside world:

Code: [Select]

INTERMUD_D: socket closing!
INTERMUD_D: prev: ({ OBJ(/secure/sefun/sefun), OBJ(/secure/daemon/ping) })
INTERMUD_D reloaded.
Loading object stack:
    0:"get_stack".OBJ(/secure/sefun/sefun)."get_stack"
    1:"create".OBJ(/daemon/intermud)."create"
    2:"CATCH".OBJ(/secure/sefun/sefun)."CATCH"
    3:"update".OBJ(/secure/sefun/sefun)."update"
    4:"CheckOK".OBJ(/secure/daemon/ping)."CheckOK"
    5:"heart_beat".OBJ(/secure/daemon/ping)."heart_beat"
Loading object trail: ({ OBJ(/secure/sefun/sefun), OBJ(/secure/daemon/ping) })
INTERMUD_D: SocketStat: 1
INTERMUD_D setup: ({ "startup-req-3", 5, "Netheria", 0, "*i4", 0, 0, -1, -1, 3000, 3005, 3008, "Dead Souls 2.4.2", "Dead Souls 2.4.2", "MudOS
    v22.2b14-dsouls2", "LPMud", "mudlib development", "mujo@net.nl", ([ "rcp" : 2990, "who" : 1, "emoteto" : 1, "http" : 2995, "channel" : 1, "finger" :
    1, "tell" : 1, "auth" : 1, "mail" : 1, "ftp" : 2999, "locate" : 1, "oob" : 3005 ]), ([ ]) })
...this above piece of code continuously stopping and starting is considered pinging right?

Code: [Select]

> mudlist -n fron
No MUDs match your query.
> mudlist -n dead
No MUDs match your query.
> mudlist -n dead*
No MUDs match your query.
>


[cratylus]

A good way to test is to change your mud's name to something
you know for sure is completely unique, and reboot. A common
problem is that if your IP changed and you reinstalled a new
copy of the lib, but you use your old mud name, the router will
assume you are not who you say you are, and reject your connection.

-Crat

[Atomic]

You're right, changed to NetheriaX and switched routers back to Yatmim, did the trick.
Scared a little at first due to the ansi-stuff scrolling by, but that was probably just to
retrieve all other router-info (the other muds).

You've stated it in the FAQ as well, I should have searched better. 

On a side-note: any changes that names get purged after a while to get the original name I
chose reconnected to a new ip or does that have to be done manually by the Yatmim-admin?

[cratylus]

The "reserved" status of mud names is supposed to drop after 10 days or so of
a mud not reconnecting. However, send me a tell sometime tonight and I'll reset
Netheria for you.

-Crat

[cratylus]

I'll be upgrading the libs on the primary and secondary routers today and tomorrow.

You may experience occasional service disruptions. Please bear with me.

If you cannot connect at all, even after switching to the alternate router,
please send me a PM to let me know.

Thanks.

-Crat

[Atomic]

The "reserved" status of mud names is supposed to drop after 10 days or so of
a mud not reconnecting.

Ah, so there is a timer-something, good to know.

However, send me a tell sometime tonight and I'll reset
Netheria for you.

Not that life-threatening at the moment, but thanks.

[cratylus]

I've upgraded both *yatmim and *i4 and fixed various IRN problems.

I think it should work just fine, but you never know. Please post if you
have connection or communication problems.

-Crat

[cratylus]

Looks like the upgrade somehow clobbered non-default channels and I didn't notice.

They should be ok now...please let me know if they're still hosed.

-Crat

[cratylus]

I've been running into some troubles with the *i4 router.

I'm working on them right now, and hope to have them resolved soon.

*i4 is the twin node of *yatmim. Normally it doesn't matter which you connect
to, you are on the same network.

However, if you're connected to *i4, you might experience some lack
of access to intermud today. My apologies for the inconvenience. If this
is problematic for you, please switch to *yatmim instead.

On a *totally* unrelated note, Arren's router, *adsr, is also having
some trouble. If you are willing to abide by the *yatmim router rules,
you're welcome to hang out on *yatmim til Arren resolves his issues.

-Crat

[cratylus]

Bad week for routers! *i4 is ok now, but the *yatmim computer is choking on
a bad disk, causing crashes and performance lag. I've been meaning to upgrade
that box anyway, and it looks like this is the Saturday to do it. *yatmim will
be completely unavailable today for a few hours.

Sorry for any inconvenience.

-Crat

[cratylus]

Ok, the *yatmim box has had the failing disk replaced, and was updated with the
current version of the OS, as well as generally patched and massaged and
fed some grapes and honey.

It will need another couple of reboots following further patching, which
will happen today and tomorrow, but other than that, everything should be
mostly cool. Please let me know if you're having trouble connecting.

-Crat

[cratylus]

The maintenance that began this weekend will be finishing today, 18 June.

*yatmim will reboot twice in order to implement some groovy new patches. Please forgive
the inconvenience.

-Crat

[cratylus]

Normally this thread is all about this problem or that problem.

For a change, I'm reporting that the status of the routers is good.

*yatmim and *i4 are now graced with code that allows:

- efficient sharing of IRN data
- proper handling of null or errored sockets
- implementing of new code without having to drop connections (a separate socket daemon handles connections)
- joy and happiness

Well, the joy and happiness are mine really

I've discovered that the router code is now at a point that is so stable that
downtime and problems occur not because of network/code bugs, but
because of hardware failures or because I manually intervened in something
while deep into my pints.

As long as I keep my grubby hands off the throttle and no more disk
or campus network failures occur, I expect the routers to chug along
happily for weeks or months at a stretch.

So, all is good. The router code is still undocumented, so if you're looking to start one,
you'll need to grit your teeth and digest the code on yer own, using the
specs at intermud.org to light your way. However, the upside is that the code
works real good.

-Crat

PS note that the most current router code is in the alpha release

[cratylus]

More "no news"!

Forgive me the self-congratulation. But for the history
of yatmim, announcements have almost invariably been
about bad things and failures...and it's just so nice that
such things are now so infrequent.

The *yatmim node has been up without interruption
now for: 7w 18h 12m 24s

And the *i4 node for: 10w 3d 13h 37m 17s

At the risk of jinxing things, I just wanted to do a little
public dance here to celebrate this record.

Just as a heads-up, though, I will probably be doing
some updating of how the router handles old muds. I
don't expect this new system to screw things up, but hey,
you know me. *I* might screw it up. I'll be testing this
weekend on a lab router pair to minimize the
downtime risk.

The intent is to have the router notify muds when an old mud
is removed. This involves your mud receiving a mudlist
with a data packet of: 0

I believe most i3 implementations handle this just fine, but
I thought it was worth bringing it up, in case folks have
homebrew implementations ill-equipped to handle this behavior.

-Crat