Le 29/09/2020 ? 16:24, Oliver R. a ?crit?:

Dear All,

I have a custom test that reports data from nvme disks from the command:

nvme smart-log /dev/nvme2n1 -o json

The output is processed so that the client reports the following to the server:

S3ESNX0J951626N-critical_warning : 0 S3ESNX0J951626N-temperature : 318 S3ESNX0J951626N-avail_spare : 100 S3ESNX0J951626N-spare_thresh : 10 S3ESNX0J951626N-percent_used : 92 S3ESNX0J951626N-data_units_read : 27832088 S3ESNX0J951626N-data_units_written : 93877408 S3ESNX0J951626N-host_read_commands : 180442558 S3ESNX0J951626N-host_write_commands : 916278700 S3ESNX0J951626N-controller_busy_time : 4028 S3ESNX0J951626N-power_cycles : 218 S3ESNX0J951626N-power_on_hours : 2995 S3ESNX0J951626N-unsafe_shutdowns : 86 S3ESNX0J951626N-media_errors : 0 S3ESNX0J951626N-num_err_log_entries : 0 S3ESNX0J951626N-warning_temp_time : 0 S3ESNX0J951626N-critical_comp_time : 0 S3ESNX0J951626N-temperature_sensor_1 : 318 S3ESNX0J951626N-temperature_sensor_2 : 323 S3ESNX0J951626N-thm_temp1_trans_count : 0 S3ESNX0J951626N-thm_temp2_trans_count : 0 S3ESNX0J951626N-thm_temp1_total_time : 0 S3ESNX0J951626N-thm_temp2_total_time : 0 2J4520102682-critical_warning : 0 2J4520102682-temperature : 314 2J4520102682-avail_spare : 100 2J4520102682-spare_thresh : 32 2J4520102682-percent_used : 2 2J4520102682-data_units_read : 9450966 2J4520102682-data_units_written : 24105094 2J4520102682-host_read_commands : 137338588 2J4520102682-host_write_commands : 284702582 2J4520102682-controller_busy_time : 0 2J4520102682-power_cycles : 46 2J4520102682-power_on_hours : 1107 2J4520102682-unsafe_shutdowns : 10 2J4520102682-media_errors : 0 2J4520102682-num_err_log_entries : 0 2J4520102682-warning_temp_time : 0 2J4520102682-critical_comp_time : 0 2J4520102682-thm_temp1_trans_count : 0 2J4520102682-thm_temp2_trans_count : 0 2J4520102682-thm_temp1_total_time : 0 2J4520102682-thm_temp2_total_time : 0 S3ESNX0J951635M-critical_warning : 0 S3ESNX0J951635M-temperature : 310 S3ESNX0J951635M-avail_spare : 100 S3ESNX0J951635M-spare_thresh : 10 S3ESNX0J951635M-percent_used : 92 S3ESNX0J951635M-data_units_read : 32693378 S3ESNX0J951635M-data_units_written : 95742837 S3ESNX0J951635M-host_read_commands : 213266959 S3ESNX0J951635M-host_write_commands : 918085461 S3ESNX0J951635M-controller_busy_time : 4280 S3ESNX0J951635M-power_cycles : 218 S3ESNX0J951635M-power_on_hours : 3072 S3ESNX0J951635M-unsafe_shutdowns : 86 S3ESNX0J951635M-media_errors : 0 S3ESNX0J951635M-num_err_log_entries : 1 S3ESNX0J951635M-warning_temp_time : 0 S3ESNX0J951635M-critical_comp_time : 0 S3ESNX0J951635M-temperature_sensor_1 : 310 S3ESNX0J951635M-temperature_sensor_2 : 320 S3ESNX0J951635M-thm_temp1_trans_count : 0 S3ESNX0J951635M-thm_temp2_trans_count : 0 S3ESNX0J951635M-thm_temp1_total_time : 0 S3ESNX0J951635M-thm_temp2_total_time : 0

As you can see, there are three disks, that have the same metrics with different values. Now I started with a xymonserver.d/nvme.cfg looking like this:

TEST2RRD="$TEST2RRD,nvme=ncv" SPLITNCV_nvme="*:GAUGE" GRAPHS_nvme="nvmecriticalwarning,nvmetemperature,nvmeavailspare,nvmesparethresh,nvmepercentused,nvmedataunitsread,nvmedataunitswritten,nvmehostreadcommands,nvmehostwritecommands,nvmecontrollerbusytime,nvmepowercycles,nvmepoweronhours,nvmeunsafeshutdowns,nvmemediaerrors,nvmenumerrlogentries,nvmewarningtemptime,nvmecriticalcomptime,nvmetemperaturesensor1,nvmetemperaturesensor2,nvmethmtemp1transcount,nvmethmtemp2transcount,nvmethmtemp1totaltime,nvmethmtemp2totaltime"

This causes all rrd files beeing created correctly like this:

$ ls -1 /var/lib/xymon/rrd/wsrbreb/nvme*temperature.rrd /var/lib/xymon/rrd/wsrbreb/nvme,2J4520102682_temperature.rrd /var/lib/xymon/rrd/wsrbreb/nvme,S3ESNX0J951626N_temperature.rrd /var/lib/xymon/rrd/wsrbreb/nvme,S3ESNX0J951635M_temperature.rrd

Unfortunately all datasets are now saved as datatype "GAUGE", but most need to have "DERIVE", but I cannot figure out how to do this. Here is what I've tried:

SPLITNCV_nvme="temperture:GAUGE" SPLITNCV_nvme="*temperture:GAUGE" SPLITNCV_nvme=".*temperture:GAUGE" SPLITNCV_nvme="%.*temperture:GAUGE"

One thing that does work is the following:

SPLITNCV_nvme="S3ESNX0J951626N_temperature:GAUGE"

But as you can see it requieres the name of the SSD, which cancels out all dynamics.

How can I work with wildcards in SPLITNCV scenarios?

Thank you for your help!

Regards

---

Xymon mailing list Xymon at xymon.com http://lists.xymon.com/mailman/listinfo/xymon

Ok Oliver,

I understand your point now: you want to use wildcard related to the dataset name. Hence, we have to validate this is possible... I could not find anything pointing to wildcard usage in define the dataset name, in this documentation: https://xymon.sourceforge.io/xymon/help/howtograph.html

I'm not a C guy, hence I'll consider documentation only: if there is no mention, it does not exist.

You could try to handle manually (awful answer I know) the serial number of your NVMe's ? Or better: contribute to the code (answer from someone lazy who won't)