NFS Client RPCs Test

This test reports the statistical information about the Connection and Connectionless RPC calls made by the NFS client. The test is applicable to Solaris OS only.

Target of the test : NFS on Solaris Client

Agent deploying the test : An internal agent

Outputs of the test : One set of results each for connection and connectionless RPC calls.

Configurable parameters for the test
Parameter	Description
Test period	How often should the test be executed
Host	The host for which the test is to be configured.

Measurements made by the test
Measurement	Description	Measurement Unit	Interpretation
Total number of calls	The total number of RPC calls made by the client during the last measurement period	Number
Number of badcalls	The number of times that an RPC call failed due to an error such as a timeout or an interrupted connection during the last measurement period.	Number	A non-zero value indicates timeouts or retransmissions. If a server has crashed, bad calls can be expected to happen. But, if bad calls happen during normal operation, then soft-mounted file systems use larger timeo value or a larger retrans value to avoid RPC failures. Note that on soft-mounted file systems, a request is retransmitted a limited number of times before it is reported as a failed RPC call. The value of badcalls is only incremented for the final failed attempt; previous failures increase the value of retrans. All requests that fail due to a timeout are recorded in timeouts.
Number of badxids	The number of responses from servers for which the client has already received a response.	Number	If a client does not receive a response to a request within a time period, it retransmits the request. It is possible that the server may service the original request. In such a case, the client receives more than one response to a request. The value of badxid is incremented for every unexpected response. If the value of badxid is approximately equal to retrans, one or more servers probably cannot service client requests fast enough. Increase the timeo parameter for the NFS mount to alleviate request retransmission or tune the server to reduce the average request service time. With a large timeout count, if badxids are reported, it indicates that the network is dropping parts of NFS requests or replies. Reduce the NFS buffer size using the rsize and wsize mount parameters to reduce the probability of NFS buffer corruption during transmission.
Number of timeouts	The number of calls that timed out waiting for response from a server during the last measurement period.	Number	If greater than 5% of all calls timeout, either the requests are not reaching the server or the timeout setting is too low. Check the badxids value to find the reason for timeouts.
Number of newcreds	The number of times authentication information had to be refreshed during the last measurement period.	Number
Number of badverfs	The number of times the call failed due to a bad verifier in the response. This is a maintenance command.	Number
Number of timers	The number of times the calculated time-out value was greater than or equal to the minimum specified timeout value for a call. This is a maintenance command.	Number
Number of cantconn	The number of requests made by the client that could not connect to the server during the last measurement period. This is specific to connection based RPC calls.	Number	If greater than 1% of the total calls cannot connect, there is usually an NFS problem. Often, this is because the NFS server is down. It can also indicate that the connection queue length in the NFS server is too small, or that an attacker is attempting a denial of service attack on the server by clogging the connection queue. If the queue length is too small, use the –l parameter to nfsd to increase the queue length.
Number of nomem	The number of times the call failed due to a failure to allocate memory. This is a maintenance command.	Number
Number of interrupts	The number of interrupted requests to a server by a client. This is specific to connection based RPC calls.	Number
Number of retrans	The number of repeated requests by the client to the server. This is specific to connectionless RPC calls.	Number
Number of cantsend	The number of requests that could not be sent by client to the server. This is specific to connectionless RPC calls.	Number