Sponsored Post: Fundamentals of VoIP Monitoring and Troubleshooting, Part 4
It's important to watch traffic entering and exiting interfaces where there is limited bandwidth and possible contention with lower priority traffic.
By Chris Lee, SolarWinds Technical Product Marketing Manager
We recently held a webcast covering Monitoring & Troubleshooting VoIP Call Quality. In case you missed it, here are a few of the great questions we received during the live session. You can also view the recorded webcast here.
Question: Is it possible to monitor QoS and utilize this information when troubleshooting call quality issues?
Answer: Absolutely, and this is a very important statistic to watch. Any real-time application like VoIP is going to be sensitive to delay and can be immediately noticed from an end-user's perspective.
QoS, like many other network statistics, can be collected from network devices along the path, using SNMP. It's important to watch traffic entering and exiting interfaces where there is limited bandwidth and possible contention with lower priority traffic, like Internet traffic. You want to look at the policy maps and ensure that traffic tagged for VoIP does not have drops during the time frame the trouble exists. If there is heavy utilization on the link and you're seeing drops, it's very possible that you need to either add bandwidth to the link or adjust your QoS policy.
Question: Can you monitor Call Detail Records and IPSLA operations that display quality metrics side by side, and what is the difference between the two?
Answer: Sure, you can monitor both at the same time with the right tool in place, but it's important to understand what you're looking at when gauging the statistics. In a typical environment, CDR/CMRs are displaying call quality metrics like jitter, latency, and packet loss from real calls and from ear to mouth between two VoIP phones.
IPSLA operations are generating synthetic traffic and sending it between two IPSLA devices, i.e. two MPLS routers, and measuring responsiveness. While this is a great way to baseline your environment and gauge call quality across the WAN, I would suggest watching both so that you can better isolate the problem.
Question: Is there a best practice for monitoring a group of remote sites and proactively identifying problems that appear to affect multiple users?
Answer: Yes, this is possible and what I would consider absolutely critical from a monitoring perspective. Most tools used for troubleshooting do not see the forest for the trees, and what appears to be a string of complaints can quickly turn into a site-wide problem. Using Call Detail Records, you can see the region or site identified with the extensions reporting problems. With the right filters in place, you can set alerts that look for an abnormal amount of failed calls or degraded call quality. This can be crucial when a site-wide problem exists, and it also helps you quickly resolve the issue.
For more information, you can also download our white paper, "Fundamentals of VoIP Monitoring & Troubleshooting" here.