Diagnostics for the LSF
rate
limiter and
lsfproxyd daemon
To troubleshoot the rate
limiter and its
interactions with the lsfproxyd daemon, you have various diagnostic
options.
Use the query_info.queryproxylog.hostname log file for
diagnosing the lsfproxyd daemon. The log file shows who issued these requests,
where the requests came from, the data size of the query, the receiving and processing times, and
whether the request was rejected or accepted. To enable the logs, either statically configure the
ENABLE_DIAGNOSE parameter in the lsb.parms
configuration file (if not defined, the default LSF is
used), or dynamically enable the diagnostic logs by running the badmin diagnose -c
query command, as described in Enabling daemon log files for diagnostics.
lsfproxyd daemon, you can also set the debugging log
class (level) for the lsfproxyd daemon, by configuring the LSF_DEBUG_PROXYD parameter in the lsf.conf file.
For example, to set to the log level to log significant program walk steps,
run:LSF_DEBUG_PROXYD="LC_TRACE"LSF_DEBUG_PROXYD="LC_TRACE LC_EXEC"Additionally, if LSB_DEBUG_MBD="LC2_LSF_PROXDY" is set in the
lsf.conf file, the mbatchd daemon captures information about
the rate
limiter
lsfproxyd daemon requests received without a
token, and logs them at the LOG_WARNING level. These warnings include the user, the
request operation code, the LSF header
version, and the host from where the request originated. This can be useful information for
identifying the user of old binaries.