[PennMUSH] Blasting my MUSH back to the stone age?

[Jeff Ferrell]

I think I'm getting some junk in my startup sequence somehow. Here's  
the deal:

Cleaned out the data folder, started with an indb.Z file from a good  
(but old!) backup. When I restart, I end up with a minimal DB.

Here's my log output (with some bits removed, for brevity's sake!):

[07/24 01:02:11] 1.8.1
[07/24 01:02:11] MUSH restarted, PID 1379, at Thu Jul 24 01:02:11 2008
Usage: sh [--port=PORT] --file=FILE

sh: error: no such option: -c
[07/24 01:02:11] Couldn't read data/indb! Creating minimal world.
[07/24 01:02:11] Reading access.cnf
[07/24 01:02:11] Seeded RNG from /dev/urandom
[07/24 01:02:11] Reading /Volumes/whitenile/users/jeff/mush/btb/game/ 
mush.cnf

[07/24 01:02:11] Reading alias.cnf
[07/24 01:02:11] Reading restrict.cnf

[07/24 01:02:11] CONFIG: directive 'max_aliases' missing from cnf  
file, using default value.
[07/24 01:02:11] CONFIG: directive 'channel_flags' missing from cnf  
file, using default value.
[07/24 01:02:13] Spawning info slave using socketpair, pid 1384, ident 1
[07/24 01:02:13] Accepted info slave from unix-domain socket
Listening on port 4201 using IPv6.
[07/24 01:02:13] 10234 file descriptors available.
[07/24 01:02:13] RESTART FINISHED.
bind failed (Possibly harmless): Address family not supported by  
protocol family
bind failed (Possibly harmless): Address family not supported by  
protocol family
bind failed (Possibly harmless): Address family not supported by  
protocol family
bind failed (Possibly harmless): Address family not supported by  
protocol family
bind failed (Possibly harmless): Address family not supported by  
protocol family
bind failed (Possibly harmless): Address family not supported by  
protocol family


NOW. Here's where it gets crazy. I was messing around with some  
scripts in OS X's Automator tonight, that had to do with restarting  
the MUSH, one was a Python script that was generating an error:  
"Usage: sh [--port=PORT] --file=FILE" -- and there it is now in my  
netmush logs.

Here's where it gets crazier: I shut everything down, restarted my  
computer, replaced my ENTIRE MUSH INSTALL with a backup from a  
DIFFERENT HARD DRIVE, and still got the same result (above) on startup.

Also, in my process list I've got:

  224 ttys000    0:43.08 ./netmush /Volumes/whitenile/users/jeff/mush/ 
btb/game/mush.cnf
  225 ttys000    0:00.00 (Python)
  226 ttys000    0:00.01 ./info_slave 12 5 1

That (Python) thing is, well, I don't know -- I'm not RUNNING anything  
Python right now.

Any ideas WHAT THE HECK I might have done to my system? (The python  
script I was trying to start through AppleScript is from Adobe, and  
has to do with serving policies for network connections in Flash 9: http://www.adobe.com/devnet/flashplayer/articles/socket_policy_files.html)

Where O where could that cruft be coming from, assuming that's what's  
breaking my restarts?

Jeff.


On Jul 24, 2008, at 12:29 AM, Talvo Hornblower wrote:

> As far as I know, the MUSH would only create the minimal db if it  
> couldn't find your db at all. Corruption should cause it to throw an  
> error and just shut down again. One possibility is that you didn't  
> restart your db with the restart script, but ran netmud directly -  
> the restart script renames "outdb" to "indb", and it's "indb" that  
> the MUSH looks for when it first boots up. Another possibility is  
> that you changed the compression type in mush.cnf, so the MUSH was  
> looking for "indb.gz" but only found "indb". But check in game/data  
> for basically any file that doesn't just contain a minimal db. game/ 
> save is also a good place to look.
>
> Mike
>
> ----- Original Message ----- From: "Andrew Brockert" <andrew at mercuric.net 
> >
> To: <pennmush at pennmush.org>
> Sent: Thursday, July 24, 2008 5:23 AM
> Subject: Re: [PennMUSH] Blasting my MUSH back to the stone age?
>
>
>> The initial blank DB is generated by the code, so yes, it's quite
>> possible. You probably did something that corrupted the DB to the  
>> point
>> where Penn couldn't read it, OR broke the DB-reading code. Either  
>> way,
>> it throws its hands up and starts fresh. This is why backups are a  
>> Good
>> Thing.
>>
>> Jeff Ferrell wrote:
>>> Hi, everybody!
>>>
>>> Interesting thing today -- I was doing some serious violence to my
>>> MUSH (shutting down, restarting, occasionally PANIC-ing, etc), and  
>>> all
>>> of a sudden, I couldn't log in. I checked my database file, and it's
>>> back to the initial "blank" DB -- one room, and the "one" character.
>>> Is that even *possible*?! I've got some old backups, but I'd really
>>> like to find my DB.
>>>
>>> Thanks!
>>> Jeff.
>>> _______________________________________________
>>> PennMUSH mailing list
>>> PennMUSH at pennmush.org
>>> http://www.pennmush.org/mailman/listinfo/pennmush
>>> PennMUSH FAQ-o-matic! http://www.pennmush.org/cgi-penn/fom
>>>
>>
>> _______________________________________________
>> PennMUSH mailing list
>> PennMUSH at pennmush.org
>> http://www.pennmush.org/mailman/listinfo/pennmush
>> PennMUSH FAQ-o-matic! http://www.pennmush.org/cgi-penn/fom
>