Google Kernel - Sudden Increase in Data Read Operations

I support the blog application ( blog.wokanxing.info , this is in Chinese) for myself, which was built on a Google application. It was like two or three years since the first deployment, and I have never encountered a single problem with quotas, because its simplicity and a small count of visits.

However, since the beginning of last month , I noticed that from time to time the application reported 500 server errors, and in the admin panel it shows a mysterious fast consumption of a free quota of datatstore data reading operations. About 10% of the free quota for reading is consumed within one hour (~ 5 thousand operating systems) , but I rely on only a dozen requests, which include data for reading data, 30 vertices, which means an average of 150 to 200 read op per request that seems impossible to me.

I have not made any changes to my codebase for several months , and I do not see any changes to the data warehouse policy or quotes. Despite this, it also confuses me how such consumption can be made. I often use memcache , which leaves the first page to the biggest player who retrieves the first threads with Post.all.order('-date').fetch(10, offset). Another query just retrieves one model with Post.get_by_key_nameand iterates through post.comment_set.

Sorry for my poor English, but can anyone give me some clues? Thank.

+3
source share
1 answer

. , .

, /-. "" "" "" (, ...). , , "robot". . -?. , GAE "robot".

, IP-, bot/bots. GAE .

0

All Articles