Continuing the discussion from Multiple Zope instances == slow buildout:
I see some people have been successfully using Memcached to share memory among Zope instances ZODB caches (I guess).
is that complicated/difficult to implement?
can somebody share that knowledge with the rest of us?
We took the code from plone.org https://github.com/plone/Products.PloneOrg/blob/master/src/Products/PloneOrg/caching.py and it works nicely for us too.
thanks! it seems to be also covered in the RAM Cache section of the documentation, I'll take a look later.
meanwhile we seem to have a small issue: the source code of Products.MemcachedManager is missing from GitHub.
can someone (@mauritsvanrees? ;)) import it so we can continue giving it proper maintenance?
I don't have the code anymore for importing from the collective to git. It was an ancient VirtualBox image with too much data taking up far too much space on my disk.
Easiest would be to take the code of the latest release and commit it to a fresh repository.
I haven't used MemcachedManager.
252.891 revisions later, one failed attempt and after many hours I was able to import the trunk into GitHub:
I just discarded the tags and branches; but at least we have the full history.
BTW, I used git-svn for this task; you can install it on Debian based systems using sudo apt install git-svn.
@hvelarde are you using MemcachedManager in production somewhere ?
no.