env bash
nvidia_gpu_: + power consumption (nvidia_gpu_power)
Category Tree: reduce number of categories
Merge pull request #553 from vlajos/typofixes-vlajos-20141204
typofixes - https://github.com/vlajos/misspell_fixer
Suspicious sections:plugins/mail/postfix_mail_stats recieved.labelplugins/nginx/nginx_vhost_traffic
Revert "Avoid bashism"
Merge pull request #534 from asgh/master
Avoid bashism
USe /bin/bash as interpreter for bash script
Merge pull request #525 from leeclemens/patch-3
Use case insensitive grep for Temp lines
backwards compatible grep for memory
Potentially will work for 5.5 and earlier, as well as 6 and 6.5 (tested with 6.5)
xorg-x11-drv-nvidia-340.29-1.el6.x86_64 (CUDA 6.5?) seems to have changed the Temperature lines from "Gpu" to "GPU" prefixes. Case insensitive grep preserves backwards comatability.
Reduce number of categories
Fix memory usage check
Change "FB Memory Usage" to just "Memory Usage" - this will also work on an old version that still uses the "FB" terminology.
_mem should use FB memory used, only
NVIDIA 6 added differentiation between FB and BAR1 memory - use nvidia_gpu_mem for FB Memory only since its definition fits better with the usage prior to BAR1 memory.
Merge pull request #430 from leeclemens/master
append newline to support _mem and multiple GPUs
Memory used percentages were being concatenated into a single string if multiple GPUs were present, e.g. both using 30% would print mem0.value 3030 mem1.value
Merge branch 'pr/401'
Merging a rebased branch for pull #401
Improve GPU plugins documentation
Configurable warning and critical temperatures for GPUs
Add TODO and FIXME notes on some issues
TODO 1: Follow multigraph suggestion from Flameeyes to look into multigraphplugins (http://munin-monitoring.org/wiki/MultigraphSampleOutput),in order to reduce the amount of round trips to get the data.
TODO 2: Put warning and critical as vars in config with sensible defaults....
deletet unnecessary output "\n"
Output looked like this:
root@htpc:/etc/munin/plugins# ./nvidia_gpu_memmem0.value 0\n
Add plugins for monitoring NVIDIA and AMD GPUs
nvidia_smi: Get memory usage even if nvmlDeviceGetUtilizationRates is not available
- have some dirs
