As strange as this may sound, I have the exact same code published on a US-EAST server and on a EU server.
Both server have billing enabled.
Th expiration is set to 7200 seconds.
memcache.set(memcache_key, result_dic, 7200)
And I have cron jobs enabled to rerun the line of code above to refresh the cache for the unique key.
- description: Cache 0:20 hour
url : /worker/topnews-cache-update
schedule: every day 0:20
- description: Cache 2:20 hour
url : /worker/topnews-cache-update
schedule: every day 2:20
This works perfectly fine for the US-EAST server.
Memcache dashboard shows 2,064 items in cache.
Memcache dashboard on the EU server however looses the objects very fast. In a short amount of time the objects disappear back to zero, despite the fact that the cron jobs keep running successfully.
It is the same code deployed on the two servers, why is memcache acting like this?
Thanks,
Houman