[Moin-user] need server-busting test

[Thomas Waldmann]

> A virtual server in itself wouldn't be so bad, but this one will be
> static -- it will have about half the memory and I/O bandwidth and
> capacity of the standalone Linux workstation we had envisioned.  This
> doesn't make a lot of sense to me, since it seems like we're going
> backwards from what we have now.
> 
> So I'm trying to come up with a test case we can use to break the
> server, in order to show that their assessment of our requirements is
> incorrect.
> 
> I have 5 or 6 wikis hosted on this server.  Generally, they don't get
> a lot of traffic, but that could be arranged.
> 
> Can anybody think of some set of wiki operations that would overload
> 4GB of memory?

I think going beyond 4GB is maybe not that easy (except if you have many
persistent worker processes running for many requests for wikis having
lots of pages - moin tries to cache some informations about pages in
memory).

So the I/O bandwidth could be more likely the limiting factor (moin does
quite much filesystem I/O), e.g. for:
 * fulltext search (not xapian search)
 * WordIndex
 * SystemInfo
 * EventStats and related (if you have a big event-log)
 * OrphanedPages

If that isn't enough, that graphviz dot and neato graph drawing
extension can consume quite some cpu (and memory iirc).

BTW, if putting it on the mainframe has to do with availability of a
critical resource (your wikis), you could also offer to run the
production wiki on a separate server, but to rsync it to a fallback wiki
on the mainframe regularly. You could even have that wiki online in
read-only mode all the time (maybe stop before rsync and start
afterwards).