On Tue, May 24, 2011 at 06:18:17PM +0200, Stephen Shirley wrote:
> On Tue, May 24, 2011 at 11:05, Iustin Pop <[email protected]> wrote:
> > As Ganeti matures, one of missing functionality is the ability to query
> > it for things like: “What happened to this instance three days ago?”,
> > “When was this instance powercycled?”, or “How many times was this
> > OpCode executed in the past week?”. The lack of such background
> > information hampers debugging of both the code and of the state of the
> > fleet.
> 
> Agreed, this is badly needed. In my case, i most frequently run into
> it when i find an instance in ADMIN_down, but have no clue as to
> *why*.
> 
> 
> > Storage details
> > ~~~~~~~~~~~~~~~
> >
> > TO BE DONE
> >
> > Needs to be replicated. Needs to be time-limited or space-limited.
> >
> > Since the size of the audit log can be significant (e.g. 1000
> > instances * 2 operations per day * 365 = 730K entries. 1M entries with
> > valid fields make a Python interpreter eat ~400MB RSS, and when
> > converted to JSON it takes 200MB), we must allow download/expire on old
> > entries. This means that long-term storage of audit logs is off-cluster.
> >
> >
> > User interface
> > ~~~~~~~~~~~~~~
> >
> > The basic functionality is, of course, examining audit logs and being
> > able to search them. Multiple ways of searching will be provided:
> >
> > - search by entity name
> > - search by UUID
> > - timestamp searches
> 
> I'm wondering why use JSON

Ah, JSON was just an example.

> , and why provide an interface? I'd be
> perfectly happy with a plain text file with one line per entry that i
> can just grep.

The issue is that this definitely needs to be replicated. Do you think
the replication should be just something alike to each node syslogg-ing
to each other node?

thanks,
iustin

Reply via email to