[checkmk-commits] Check_MK Git: check_mk: esx_vsphere_counters.diskio: now also shows disk latency

Module: check_mk Branch: master Commit: c2ab83d35666ffab99e551a3fc8bb692ddd4dcf4 URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=c2ab83d35666ff… Author: Andreas Boesl <ab(a)mathias-kettner.de> Date: Mon Feb 17 16:20:54 2014 +0100 esx_vsphere_counters.diskio: now also shows disk latency --- .werks/556 | 8 ++++++++ ChangeLog | 1 + checks/esx_vsphere_counters | 9 ++++++--- 3 files changed, 15 insertions(+), 3 deletions(-) diff --git a/.werks/556 b/.werks/556 new file mode 100644 index 0000000..378740a --- /dev/null +++ b/.werks/556 @@ -0,0 +1,8 @@ +Title: esx_vsphere_counters.diskio: now also shows disk latency +Level: 1 +Component: checks +Version: 1.2.5i1 +Date: 1392650420 +Class: feature + + diff --git a/ChangeLog b/ChangeLog index b66d930..1931758 100644 --- a/ChangeLog +++ b/ChangeLog @@ -90,6 +90,7 @@ * 0701 emc_isilon_diskstatus: new check for Status of Disks in EMC Isilon Storage Systems * 0581 emc_datadomain_disks emc_datadomain_fs: new checks to monitor disks and filesystems of EMC Datadomain * 0718 logwatch.ec: Optionally monitor the list of forwarded logfiles... + * 0556 esx_vsphere_counters.diskio: now also shows disk latency * 0103 FIX: services: Fixed bug with service inventory defined in main.mk... * 0299 FIX: borcade_mlx_fan: Prettified output, handling "other" state now * 0300 FIX: cisco_fru_power: Trying not to inventorize not plugged in FRUs... diff --git a/checks/esx_vsphere_counters b/checks/esx_vsphere_counters index 58f7385..89941be 100644 --- a/checks/esx_vsphere_counters +++ b/checks/esx_vsphere_counters @@ -77,6 +77,7 @@ def inventory_esx_vsphere_counters_diskio(info): def check_esx_vsphere_counters_diskio(_no_item, _no_params, info): reads = 0 writes = 0 + latency = 0 for counter, item, value, unit in info: if item == '': if counter == 'disk.read': @@ -87,10 +88,12 @@ def check_esx_vsphere_counters_diskio(_no_item, _no_params, info): reads += int(value) elif counter == 'disk.numberWrite': writes += int(value) + elif counter == 'disk.deviceLatency': + latency = max(latency, int(value)) - return 0, "%s/sec read, %s/sec write, IOs: %.2f/sec" % ( - get_bytes_human_readable(read_bytes), get_bytes_human_readable(write_bytes), reads + writes,), \ - [ ("read", read_bytes), ("write", write_bytes), ("ios", reads + writes) ] + return 0, "%s/sec read, %s/sec write, IOs: %.2f/sec, latency: %d ms" % ( + get_bytes_human_readable(read_bytes), get_bytes_human_readable(write_bytes), reads + writes, latency), \ + [ ("read", read_bytes), ("write", write_bytes), ("ios", reads + writes), ("latency" , latency) ] check_info['esx_vsphere_counters.diskio'] = {