Specifications
LifeKeeper Flags
path in the DEAD state, which initiates a failover event if there are no other
communications paths marked ALIVE.
l Remote system writer process transmits a LifeKeeper maintenance message, causing
the reader process to perform the protocol necessary to receive the message.
l #NAKs. Number of times the writer process received a negative acknowledgment (NAK). A
NAKmessage means that the reader process on the other system did not accept a message
packet sent by the writer process, and the writer process had to re-transmit the message
packet. The #NAKs statistic can accumulate over a long period of time due to line noise. If,
however, you see the numbers increasing rapidly, you should perform diagnostic procedures
on the communications subsystem.
l #chksumerr. Number of mismatches in the check sum message between the servers. This
statistic can accumulate over a long period of time due to line noise. If, however, you see the
numbers increasing rapidly, you should perform diagnostic procedures on the communications
subsystem.
l #incmpltmes. Number of times the incoming message packet did not match the expected
size. A high number of mismatches may indicate that you should perform diagnostic
procedures on the hardware port associated with the communications path.
l #noreply. Number of times the writer process timed out while waiting for an acknowledgment
and had to re-transmit the message. Lack of acknowledgment may indicate an overloaded
server or it can signal a server failure.
l #pacresent. Number of times the reader process received the same packet. This can happen
when the writer process on the sending server times out and resends the same message.
l #pacoutseq. Number of times the reader received packets out of sequence. High numbers in
this field can indicate lost message packets and may indicate that you should perform
diagnostic procedures on the communications subsystem.
l #maxretrys. Metric that increments for a particular message when the maximum
retransmission count is exceeded (for NAK and noreply messages). If you see a high number
in the #maxretrys field, you should perform diagnostic procedures on the communications
subsystem.
LifeKeeper Flags
Near the end of the detailed status display, LifeKeeper provides a list of the flags set for the system.
A common type is a Lock LCD flag used to ensure that other processes wait until the process lock
completes its action. The following is the standard LCD lock format:
!action!processID!time!machine:id.
These are examples of general LCD lock flags:
l
!action!02833!701236710!server1:filesys. The creation of a file system hierarchy produces a
flag in this format in the status display. The filesys designation can be a different resource type
for other application resource hierarchies, or app for generic or user-defined applications.
SteelEye Protection Suite for Linux53