[Zope-CMF] backup strategies

sean.upton@uniontrib.com sean.upton@uniontrib.com
Fri, 07 Feb 2003 13:03:05 -0800

Thanks for the clarification.  I haven't had a chance to use the new
replication tool yet.  I'd be interested to see if ClientStorage has
problems in this area.  I've done failover with heartbeat this way on a ZEO
1 cluster many times, but likely never when live writes were happening to
the storage...


-----Original Message-----
From: Toby Dickenson [mailto:tdickenson@geminidataloggers.com]
Sent: Friday, February 07, 2003 12:36 PM
To: sean.upton@uniontrib.com; pw_lists@slinkp.com; zope-cmf@zope.org
Subject: Re: [Zope-CMF] backup strategies

On Friday 07 February 2003 6:59 pm, sean.upton@uniontrib.com wrote:

> Heartbeat also manages resource with init-like scripts, when the takeover
> started, it would start up a ZSS process on a replicated DirectoryStorage
> after taking over the IP.  For safety, you would want to likely kill the
> primary server to keep it from replicating to the backup after the
> takeover. You could do this by using a power-device (STONITH: Shoot The
> Other Node In The Head).

STONITH shouldnt be necessary for safety if using DirectoryStorage's new 
replication tool. File locking takes care of preventing a replication taking

place while the replica storage is live, and revision checking in the 
replication tool will forever block replication if a write transaction 
happens inside the live replica, even once the file locks have gone.

Is ClientStorage reliably safe in this context? I suspect it may be possible

for it to get confused if the replica storage comes up missing the last 
couple of trasactions. I have been planning some experiments in this area. 
There is an easy solution to any problem; ClientStorage needs to trigger a 
full Zope shutdown and restart when it loses its connection to the ZEO 

> > Is there any way for example to *test* Data.fs when you back it up (to
> > sure that you are not backing up corrupt data)?
> there's a utility in utilities/ZODBTools/fstest.py that checks
> for errors. Run a cron job that runs this tool and mails you the
> result.
> There is also another utility, fscheck.py that gives more extensive
> reports, and IIRC is new to Zope 2.6.

And for DirectoryStorage, checkds.py

Toby Dickenson