The Client Access Server (CAS) is dedicated to user connectivity in the Exchange environment, ensuring mailbox access through various protocols. The health of this critical bridge depends on multiple services.
Within any company, users must be able to access their mailboxes through their mobile phone (ActiveSync), remote connection and web browser (OWA), and their normal clients (POP3 or IMAP).
GSX Monitor & Analyzer main features
Monitoring CAS critical Windows services
Monitoring of the global CAS Role status
Protocol monitoring
Threshold on OWA average response time
Threshold on download task queued
Critical WMI counters
Troubleshooting graphs
Monitoring maintenance set up
Trending CAS activity
When it comes to Client Access Connectivity, GSX helps ensure that users can connect their mailbox and their experience meet all their expectations.
Monitoring CAS critical Windows services
GSX lists the critical Windows services that must run on the CAS and alerts if one is down. Administrators can add optional non-critical Windows services to monitor.
Monitoring of the global CAS Role status
GSX alerts when a Role is down (one of the protocols is failing).
Protocol monitoring
To ensure the availability of the CAS protocols, GSX provides:
- Connectivity checks: which test various protocol connections, and give the average response time for each protocol.
- Protocol connectivity latency alerts: simulate actual CAS use by accessing a test mailbox, listing the mails available, and opening the first one it finds. These full scenario tests show the actual time it takes users to access their emails, protocol by protocol.
Threshold on OWA average response time
GSX sends an alert if the test of a protocol connection exceeds a pre-defined amount of time.
Threshold on download task queued
GSX sends an alert if one or more Outlook client is waiting for Offline Address Book synchronization (Microsoft recommends a zero threshold).
Outlook Web Services automatic checks and alerts
GSX uses several PowerShell commands to perform multiple tests of the availability of these critical services:
- Availability Service
- Offline Address Book
- RPC over HTTP
- Autodiscover
Critical WMI counters
GSX provides information on the status of critical WMI counters:
Status of the FDSOAB: Queue and Completed
- Queued: The download task queue can back up when multiple clients ask for a copy at the same time, especially when everybody opens their mailbox. If this counter explodes, the server can have a performance problem or fail to spread the OAB.
- Completed: The number of users who have downloaded the OAB successfully.
RPC Client Access User Count: provides a continuous display of the population accessing the server with the RPC protocol. This information is correlated with the RPC over HTTP response time in the Outlook Web Service part of the Statistics View.
Network Interface: shows the inbound and outbound traffic in KB/s. This information helps to detect any problem (e.g. no traffic, or explosion of the traffic) with this critical interface.
Troubleshooting graphs
GSX provides instant critical graphs on the last 3000 scans for Total User Connections and Connectivity latency per Protocol. These graphs are critical to troubleshooting when CAS is down, or when there is repeated loss of availability. Administrators have to know when the latency happens in order to correlate this information with the number of user connections, for example, or with system information.
Disk space alerts
GSX allows administrators to set up proactive storage alerts on disks attached to any Exchange server, for when the available space is below a certain number of MB or percent of the disk.
Monitoring maintenance set up
GSX enables administrators to set up monitoring maintenance windows according to time zone. During this time, GSX will not send alerts and will separate the statistics so as not to interfere with the SLAs. Two sets of statistics will be reported in GSX Analyzer: normal and maintenance.
Trending CAS activity
Exchange administrators need to trend the availability of services delivered by the CAS on a daily, weekly and monthly basis, in order to detect associated short and long-term issues.
- System Server Statistics: show the availability of the server based on the WMI connectivity and Critical Services availability (CPU, RAM, server and network availability, and access time). These statistics can be correlated with the availability of various protocols to help detecting problems.
- For each Protocol: Up, Down, and Percent Up over the past 24 hours with or without maintenance, during and outside of business hours.
- Specific CAS server statistics: Average of User connections, Average of User connections without maintenance, Average of User connections during business hour without maintenance, Maximum Concurrent User Connections, Maximum Concurrent User Connections without maintenance, Maximum Concurrent User Connections Business Hours without maintenance.
GSX calculates the average and the peak of each of the critical statistics, to see how the sizing of the Infrastructure supports its real usage. For example, if the sizing was for 500 concurrent connections, and the number of connections often exceeds 700, this can lead to latency.
