Skip to main content

Simple Testing Can Prevent Most Critical Failures: An Analysis of Production Failures in Distributed Data-Intensive Systems

Large, production quality distributed systems still fail periodically, and do so sometimes catastrophically, where most or all users experience an outage or data loss. We present the result of a comprehensive study investigating 198 randomly selected, user-reported failures that occurred on Cassandra, HBase, Hadoop Distributed File System (HDFS), Hadoop MapReduce, and Redis, with the goal of understanding how one or multiple faults eventually evolve into a user-visible failure. We found that from a testing point of view, almost all failures require only 3 or fewer nodes to reproduce, which is good news considering that these services typically run on a very large number of nodes. However, multiple inputs are needed to trigger the failures with the order between them being important. Finally, we found the error logs of these systems typically contain sufficient data on both the errors and the input events that triggered the failure, enabling the diagnose and the reproduction of the production failures.  We found the majority of catastrophic failures could easily have been prevented by performing simple testing on error handling code – the last line of defense – even without an understanding of the software design. We extracted three simple rules from the bugs that have lead to some of the catastrophic failures, and developed a static checker, Aspirator, capable of locating these bugs. Over 30% of the catastrophic failures would have been prevented had Aspirator been used and the identified bugs fixed. RunningAspirator on the code of 9 distributed systems located 143 bugs and bad practices that have been fixed or confirmed by the developers.

View the presentation here or read the full paper here.

Ron - the problem management guy


Popular posts from this blog

LDWin: Link Discovery for Windows

LDWin supports the following methods of link discovery: CDP - Cisco Discovery Protocol LLDP - Link Layer Discovery Protocol Download LDWin from here.

Battery Room Explosion

A hydrogen explosion occurred in an Uninterruptible Power Source (UPS) battery room. The explosion blew a 400 ft2 hole in the roof, collapsed numerous walls and ceilings throughout the building, and significantly damaged a large portion of the 50,000 ft2 building. Fortunately, the computer/data center was vacant at the time and there were no injuries. Read more about the explosion over at hydrogen tools here .

STG (SNMP Traffic Grapher)

This freeware utility allows monitoring of supporting SNMPv1 and SNMPv2c devices including Cisco. Intended as fast aid for network administrators who need prompt access to current information about state of network equipment. Access STG here (original site) or alternatively here .