New Message: Re: Databases corrupting

[webmaster at userland.com]

A new message was posted:

Address: http://frontier.userland.com/discuss/msgReader$14085

By: Steve Hooker (steve at cybersaps.com)

You wouldn't need to quit Frontier to close properly, move, replace and open a freshly squished root.

You could switch off threads here: user.scheduler.prefs.runThreads = false and use scheduler.monitorThreads () «kills running threads, OK as agents are disabled

And shut agents with Frontier.enableAgents (false); «shut down agents
system.verbs.builtins.inetd.stop ()«stop serving

My squish script has the above, and squishes, moves, opens and restarts threads and agents.
http://www.cybersaps.org/2004/01/Fat-bloated-files-Squish-them-with-this.html

I was having a similar problem with a 140,000 message Manila site, growing everyday. Eventually, I changed 'save every minute' to 'save every 20 minutes' and counted 60 new messages before the ordinary Manila 'save manilasite after a single new message.' After heavy activity (80,000 page builds, 600 new messages) my usual nightly squishes would take 8 minutes instead of the usual 5, just on that one ManilaWebsite.

It maybe that you have some corruption in that one root, and a squish won't rid it for you. Try working through the tables exporting, when you find one that won't, dig deeper to find the actual table. There maybe several, look all over. Once properly cleaned, I wonder if a quick squish after your big update will ease your woes and if you could do that at periods of low activity?

I think I've said before, I squish every night, and check my logs for unusual bloat, or excessive times. Then dance naked around a potted cactus, three times.

Steve Hooker

http://www.cybersaps.org/

This is a Manila site.. http://manila.userland.com/.