Skip to main content

Memcached stale keys on new code deployment and evictions

Scalability problems are interesting. This weekend we deployed a new release and suddenly we started getting ldap alerts from one datacentre. All fingers were pointed to something new in the code making more ldap queries but I didnt thought we made any ldap code changes, infact we were moving away from ldap to mysql.  We use 15G of memcached and looking into memcache stats I found that we were seeing evictions.  We had bumped memory from 12G to 15G in other data centre during same code and that data centre was not having issues. So this problem was interesting.

At last we found out that our new release had made 90% of memcached data stale. We store our objects as Json in memcached and deprecated fields removal from objects made the json in cache unusable. But the code bug was that the old jsons were still sitting in cache consuming memory.  As new jsons was getting pumped into memcached they were fighting for memory and memcached  LRU is slab based so when it evicts a slab it would evict good data also. This in turn was causing Thundering herd issue in LDAP.

LDAP is nice but when it comes to scaling more than 2-3 million objects it sucks as internally openldap uses bdb backend and after a point both read/writes start taking a long time no matter what you do. Good news is that in coming month we are moving from ldap->mysql which would make us scalable again.  I will write a blog on how to migrate millions of customers from ldap->mysql without them knowing about it. The analogy is changing tires of race cars on a race track and letting cars again on race track without disrupting the ongoing race.


Comments

Popular posts from this blog

RabbitMQ java clients for beginners

Here is a sample of a consumer and producer example for RabbitMQ. The steps are
Download ErlangDownload Rabbit MQ ServerDownload Rabbit MQ Java client jarsCompile and run the below two class and you are done.
This sample create a Durable Exchange, Queue and a Message. You will have to start the consumer first before you start the for the first time.

For more information on AMQP, Exchanges, Queues, read this excellent tutorial
http://blogs.digitar.com/jjww/2009/01/rabbits-and-warrens/

+++++++++++++++++RabbitMQProducer.java+++++++++++++++++++++++++++
import com.rabbitmq.client.Connection; import com.rabbitmq.client.Channel; import com.rabbitmq.client.*; public class RabbitMQProducer { public static void main(String []args) throws Exception { ConnectionFactory factory = new ConnectionFactory(); factory.setUsername("guest"); factory.setPassword("guest"); factory.setVirtualHost("/"); factory.setHost("127.0.0.1"); factory.setPort(5672); Conne…

What a rocky start to labor day weekend

Woke up by earthquake at 7:00 AM in morning and then couldn't get to sleep. I took a bath, made my tea and started checking emails and saw that after last night deployment three storage node out of 100s of nodes were running into Full GC. What was special about the 3 nodes was that each one was in a different Data centre but it was named same app02.  This got me curious I asked the node to be taken out of rotation and take a heap dump.  Yesterday night a new release has happened and I had upgraded spymemcached library version as new relic now natively supports instrumentation on it so it was a suspect. And the hunch was a bullseye, the heap dump clearly showed it taking 1.3G and full GCs were taking 6 sec but not claiming anything.



I have a quartz job in each jvm that takes a thread dump every 5 minutes and saves last 300 of them, checking few of them quickly showed a common thread among all 3 data centres. It seems there was a long running job that was trying to replicate pending…

Email slavery

It seems I have become an EmailSlave. The first half of the day is spent in just answering to emails. There are so many emails where I am copied but I need not be. There are many emails  where its a 1-2 page email and somewhere down someone says @KP please answer this.  So it seems daily my work schedule is:
Signin to newrelic and check anomalies for 15 min. Check emails related production exception report and yes there are a ton of these report daily. Need a better tool here as this model is not scalable. I need to reduce the incoming data at me to only see relevant data like what newrelic does. May be I need to create a webapp out of these emails.Check emails for next few minutes before team callsDo team callsThen again back to checking emails until a I have taken a best shot at answering everyone waiting for my reply.Attend team meetings on Tue/Thu
Being an architect and coder at heart I don't feel satisfied at end of the day if there is nothing tangible getting done at the end.…