Q: A server's disks are incredibly slow. `iostat -x 1` shows %util at 100%. How do you identify whether it's latency on reading data or writing data?
While %util shows the disk is completely saturated (100% time spent doing I/O), it doesn't explain the load profile.
iostat columns.. I structure my answer around systematic triage first, root cause analysis second, and permanent remediation third.""
Want to master this scenario in a live sandbox? The Linux Foundation's Certified System Administrator (LFCS) & SRE Track covers this exact problem with hands-on terminal drills.
🛠️ Production Runbook & Step-by-Step Resolution
Direct Answer: "We encountered this OS-level bottleneck during peak traffic and diagnosed it down to kernel and filesystem metrics. The interviewer is testing: Deciphering complex I/O metrics, iostat columns.. I structure my answer around systematic triage first, root cause analysis second, and permanent remediation third."
Initial Diagnostics & Root Cause Analysis
While %util shows the disk is completely saturated (100% time spent doing I/O), it doesn't explain the load profile.
r_awaitvsw_await: These denote the average wait time (in milliseconds) for Read requests vs Write requests to be served. Ifr_awaitis 5ms andw_awaitis 500ms, the disk is severely struggling to write.rkB/svswkB/s: These denote the actual volume of Kilobytes Read vs Written per second.
Remediation & Permanent Safeguards
In the iostat output, I would focus on two specific column groups: If the write latency is massive, turning on Write-Back caching on a RAID controller, or migrating to an SSD, is the immediate hardware solution.
- r_await vs w_await: These denote the average wait time (in milliseconds) for Read requests vs Wri...
- rkB/s vs wkB/s: These denote the actual volume of Kilobytes Read vs Written per second.
Senior hiring managers and bar raisers evaluate how you prioritize during outages. Avoid these common anti-patterns:
- Issuing immediate kill -9 (SIGKILL) commands rather than allowing graceful termination via kill -15 (SIGTERM) when configuring servers disks incredibly SIGKILL immediately halts execution without allowing processes to flush memory buffers, commit database transactions, or remove stale PID locks.
-
Overlooking file system inode exhaustion (df -i) when diagnosing 'No space left on device' errors when configuring servers disks incredibly
Millions of small files or unrotated session tokens can exhaust available file system inodes even when standard disk storage (
df -h) shows ample free space. - Exhausting open file descriptor allocations (ulimit -n) under intensive concurrent socket connections when configuring servers disks incredibly Reaching file descriptor exhaustion throws 'Too many open files' errors, preventing services from accepting new TCP client socket connections.
-
Permitting defunct zombie processes to accumulate without fixing parent process SIGCHLD waitpid reaping when configuring servers disks incredibly
Zombie processes consume process table entries; accumulating zombies can exhaust the system maximum PID limit (
pid_max).