Checking Error Counters on Interfaces After Every Major Production Run
After every major production run, check interface error counters using *show interfaces* to catch CRC errors, runts, giants, and overruns-each can degrade audio and video quality. Zero CRC, runts, or giants mean clean signaling; sustained overruns hint at buffer exhaustion during high-traffic streams. Correlate spikes with CPU usage and SNMP data, especially if discards jump from baseline. High overruns during live 1080p streams often point to inadequate hardware or missing QoS. You’ll see exactly how to fix persistent issues next.
We are supported by our audience. When you purchase through links on our site, we may earn an affiliate commission, at no extra cost for you. Learn more. Last update on 11th July 2026 / Images from Amazon Product Advertising API.
Notable Insights
- Use the `show interfaces` command to check real-time error counters after production runs for immediate insight into link stability.
- Monitor CRC errors, runts, and giants to identify physical or data-link layer issues post-production.
- Zero or stable CRC, runt, and giant counts indicate a healthy interface with no data corruption or misconfigurations.
- Investigate sudden spikes in input errors or overruns, as they may signal cabling faults, interference, or buffer exhaustion.
- Correlate interface error trends with CPU usage and traffic load to detect hardware limitations or oversubscription issues.
Check Interface Errors With Show Interfaces
When you’re troubleshooting network issues on Cisco gear, running the `show interfaces` command gives you immediate access to critical error counters that can pinpoint problems fast. The show interface command reveals real-time error statistics, helping you catch interface errors before they disrupt live streaming or audio/video workflows. You’ll see CRC errors-often from electromagnetic interference or bad cables-plus runts, which are packets under 64 bytes caused by collisions or hardware faults. Giants, packets over 1518 bytes, show up too, usually from misconfigurations. Clean error stats with zero CRC, runts, or giants after a production run mean your data-link layer is solid. These error statistics give you confidence in your network’s physical layer, ensuring reliable signal delivery across AV systems and streaming setups. Use this command regularly to maintain peak performance.
Common Interface Error Types Explained
You’ve seen how the `show interfaces` command pulls up real-time error stats that keep your AV workflows running smoothly, and now it’s time to break down what those specific errors actually mean. CRC errors show up when a frame’s checksum doesn’t match, usually due to interference or bad cabling, and they corrupt data integrity. Runts are frames smaller than 64 bytes, often from collisions or faulty gear, and they’re dropped automatically. Giants-frames over 1518 bytes-typically come from misconfigured NICs and can delay your stream. Input errors bundle CRC, runts, and frame alignment issues; high counts mean check cables, ports, or duplex settings. Output errors point to problems on the transmit side, often tied to hardware faults or cabling. Keeping these low guarantees clean signal timing and reliable video/audio sync across your production network.
What Overruns, Discards, and Errors Really Mean
Though your network might seem stable at first glance, overruns can still sneak in when traffic floods the interface faster than the ASA’s FIFO buffers can handle, especially during bursts of small packets with tight inter-arrival timing. When that happens, overruns spike, and you’ll see the counter rise in “show interface” output-clear signs the ASA can’t keep up. These aren’t discards from ACLs or QoS; those are intentional drops. Overruns mean packets are lost before processing. Input errors like CRC, frame, or alignment issues, on the other hand, usually point to cabling faults or EMI. Discards, tracked via SNMP as ifInDiscards, often reflect policy-based drops on firewalls. Together, these counters help you pinpoint whether problems lie in traffic volume, physical layer health, or policy logic-key for keeping live audio and video streams stable.
Recognize Expected vs. Critical Discards
Discards on your ASA aren’t always a red flag-some are part of normal firewall behavior, like when traffic hits a deny rule in your ACL and gets logged as a discard. These expected discards, often from QoS or ACL drops, won’t cause performance issues or packet loss. But when discards spike suddenly-jumping from 0 to hundreds per poll-they likely signal critical problems like bufferReceived exhaustion or FIFO overflows. Those are interface error types that demand action. Use the Top 10 Errors and Discards Report to track SNMP delta changes; sustained high discards mean ongoing issues, not just noise. In SolarWinds, check the InterfaceErrors_Detail table to spot trends over time. Real-world testing shows baseline discards under 10 per interval are typical, but anything higher risks errors and degraded throughput. Separate the normal from the dangerous by analyzing where discards occur and how they correlate with system load, not just isolated counter bumps.
Match Errors to CPU and Traffic Spikes
When you’re troubleshooting network issues on your ASA, matching interface errors to CPU and traffic spikes can quickly narrow down root causes, especially during peak production runs. You’ll want to check the output of “show interface” for increasing overrun errors on any network interface-these often spike when CPU usage surges, visible via “show proc cpu-hog”. Bursts of small packets input, common in live audio/video streams, fill NIC FIFO buffers fast, causing overruns during traffic spikes. Use SNMP data to correlate ifInErrors with CPU and traffic graphs; timing is key. The Top 10 Errors and Discards Report helps pinpoint affected interfaces using MIB counter deltas. If you see repeated errors during high packets input periods, especially with runts or CRC issues, it may signal your ASA is oversubscribed. Persistent overruns mean your hardware can’t keep up, particularly with high-frequency, low-latency streaming traffic-consider upgrading or redistributing load.
Respond to High Overruns and Discards
If you’re seeing high overruns like 2881 on GigabitEthernet0/1, your ASA’s FIFO buffers are getting swamped, and packets are hitting the floor-especially during intense live streaming sessions with small, rapid-fire audio and video packets. High overruns often mean your CPU can’t keep up, but discards might also stem from firewall rules, not just congestion. Always check error counters to rule out faulty hardware or cabling. Frequent interface resets? That’s a red flag. While flow control can help, don’t enable it blindly-verify downstream switch support.
| Issue | Possible Cause | Action |
|---|---|---|
| High overruns | CPU saturation, small packets | Monitor traffic bursts |
| High discards | ACL drops, congestion | Check firewall logs |
| Rising error counters | CRC, cable faults | Inspect physical layer |
| Interface resets | Instability, misconfig | Review logs for reloads |
| Persistent issues | Faulty hardware | Replace cables or module |
Use Flow Control and QoS to Reduce Overruns
You’ve seen how high overruns on your ASA-like that 2881 count on GigabitEthernet0/1-point to FIFO buffers being overwhelmed, especially during live streaming with tight bursts of audio and video traffic, and now it’s time to take control of the flow. Enable flow control with “flowcontrol send on” to reduce overruns; your ASA (8.2(2)+ for 10GE, 8.2(5)+ for 1GE) can then send pause frames to adjacent switches when the FIFO buffer fills. Those pause frames temporarily halt incoming packet bursts, letting switches absorb traffic with their larger buffers. While the ASA can’t receive pause frames, this one-way flow control still prevents drops during intense live production. Pair it with network-wide QoS to shape and prioritize traffic early, smoothing bursts before they hit the firewall. Together, flow control and QoS minimize overruns, keep audio/video streams stable, and protect your production quality-no guesswork, just reliable performance.
On a final note
You’ll catch issues early by checking error counters after each run, especially overruns and discards on congested links. High overruns often mean buffer saturation, common on 1G interfaces pushing 900+ Mbps. Discards tied to CPU spikes suggest routing or QoS missteps. Use hardware like Cisco Catalyst 9300s with QoS policies and flow control enabled. Real-world tests show jitter drops below 30ms when you align buffer settings with traffic patterns, keeping streams stable.
