I applied online. The process took 1 week.

Round 1 : Phone screen Technical Phone screen lasted for about an hour. Went over resume , recent projects and some basic details about the company. Also got a minor dose of advice about how to select start ups ( mostly sounding annoying & pseudo elitist ) . Also bragged a little about the founders and about the interviewer himself for a little bit.

Round 2 : Went onsite to their office at San Jose. The recruiter was really annoying and literally forced me into interviewing that day. Used cheap tactics and was unavailable during regular work hours. The recruiter had also called a few of my friends at untimely hours and bothered them. On the day of the interview , half of the team was working from home or unavailable. The recruiter also pretended to be scheduling a talk with the founder , and I learnt it was a cheap trick later and nothing like that was even planned.

The QA lead was also equally annoying , and endlessly continued to brag about himself and the founder. The worst point in the interview was , when they asked me for the other companies I was interviewing and literally spent 20 minutes badmouthing every single one of them. I was left completely surprised and couldn't even politely ask them to move on.

Round 1 Questions: ============== + What is the difference between NAS and SAN ?

• The most basic answer is that NAS is a file storage and SAN is a block storage system . NAS predominatly is either NFS or CIFS protocol(this is a microsoft only protocol , and I havent done much with this at all) ; SAN usually contains Fibre Channel / FCoE and iSCSI elements. ( though an answer like this might be sificient for a phone interview , this is more scope to elaborate )

• Basic Linux commands Questions - + How would you monitor resource consumption on linux ( and variants of this)

• commands like top , free , netstat , lsof , iostat , vmstat , ps , pstree , sar are useful for linux questions. the answers generally revolve around one of these.

Note : uptime , w and route are useful commands in this situation. If the interviewer is looking for a more complicated way of monitoring ; collectd , Nagios , Cacti , Splunk are some full blown tools for monitoring and capacity planning

• How would you test a NFS server ? ( or it was a Web Server )

Round 2 : Onsite interview ==================== Spoke to three engineers in all .

Round 2.1 : + Given a python array with numbers like [ 1,4,6,9,10,2,3,14,11] ==> How would you re-arrange it as [1-4,6,9-11,14] (ie) if the array has three or more consecutive number ; club them togather and represent as starting number - ending number ; code and test the same

• Given a file , how would you search for a particular string ?

• using Grep is a fair start

• How would you do it for multiple files in a directory ?

• How would you do it for files in multiple machines (the logs are all in the same directory /root/log ) ?

Round 2.2 : This was a pretty loaded interview .

• Explain the linux boot process ?

• The short answer to this question is that , the boot process is divided into three stages ; the BIOS stage , the Bootloaded stage and Kernel stage .

In the BIOS stage - the system does a Power On self test and makes sure all the Hardware components are working. And it then scans for the first available bootable disk.

In the Bootloader stage - The context of the system is in the boot disk . The boot disk has an MBR ( master boot recond ) and the partition table . The MBR is always on the first sector of the disk ( also called the boot sector).

There is enough code in the MBR to load the next part (ie) GRUB - Grand Unified Boot loader .

In the kernel stage - the GRUB finds the location of the OS and loads it.

link : http://askubuntu.com/questions/173248/where-is-the-bootloader-stored-in-rom-ram-or-elsewhere http://unixbhaskar.blogspot.com/2010/03/insight-into-gnulinux-boot-process.html

• In ESX , how would you find resource utilization ?

• esxtop

• What is the difference between NAS/SAN ?

• Discussed multi-pathing in SAN (as a part of major difference)

• How to monitor Disk I/O on a linux box ?

• SAR and iostats are a couple of interesting tools.

( other interesting tools could be iotop , iodump , dstat )

• or we could cat /proc/diskstats

• How would you list all devices on a Linux box ?

• What is a HBA ?

• What is a iSCSI initiator and a iSCSI target ?

• What network load generation tools did you use ? ( totally zoned out for this )

• should have said io tools exert n/w load too , since all storage were remote.

Round 2.3 : + How would you test gmail ?