Merge lp:~stub/launchpad/memcache into lp:launchpad

[Stuart Bishop]

The TALES memcache integration is showing significant performance gains... when the cache is hit. Unfortunately, individual appservers are not sharing cache because they each use an individual config file. We use the config name as part of the cache key because we don't want lpnet and edge servers sharing cached information, but do want them using the same memcached servers. Stripping the digits from the end of the config name is a quick and simple way of doing what we need - allowing edge1-4 to share cache with each other but not with lpnet1-16.

[Tim Penhey (thumper)]

review: Approve

=== modified file 'lib/lp/services/memcache/tales.py'
--- lib/lp/services/memcache/tales.py	2010-03-03 10:31:33 +0000
+++ lib/lp/services/memcache/tales.py	2010-03-17 07:45:43 +0000
@@ -109,6 +109,10 @@
         '_'*33 + ''.join(chr(i) for i in range(33, ord(':'))) + '_'
         + ''.join(chr(i) for i in range(ord(':')+1, 127)) + '_' * 129)
 
+    # We strip digits from our LPCONFIG when generating the key
+    # to ensure that edge1 and edge4 share cache.
+    _lpconfig = config.instance_name.rstrip('0123456789')
+
     def getKey(self, econtext):
         """We need to calculate a unique key for this cached chunk.
 
@@ -165,8 +169,7 @@
         # tools - it doesn't really matter provided all the components are
         # included and separators used.
         key = "pt:%s:%s,%s:%s:%d,%d:%d,%s" % (
-            config.instance_name,
-            source_file, versioninfo.revno, uid,
+            self._lpconfig, source_file, versioninfo.revno, uid,
             econtext.position[0], econtext.position[1], counter,
             sanitized_url,
             )