061b699a97
Add OpenBSD to README (Thank you @mpfz0r)
2016-11-25 09:04:14 +09:00
136f6ca669
Merge pull request #283 from mpfz0r/master
...
Add support for OpenBSD/amd64
2016-11-25 07:58:43 +09:00
b4846b445b
Add support for OpenBSD/amd64
...
This code is based on the FreeBSD version
and implements roughly the same feature set.
2016-11-24 22:30:29 +01:00
110eb1f082
Merge pull request #280 from shirou/darwin-goslices
...
cpu[darwin]: fix Go arrays panic in cgo #279
2016-11-17 11:43:14 +09:00
ae251eb6a2
cpu[darwin]: fix Go arrays panic in cgo #279
2016-11-15 00:53:20 +09:00
a63900a44b
update README about taging.
2016-10-27 22:29:33 +09:00
1516eb9ddc
Merge pull request #275 from sparrc/sparrc-issue-245
...
Fix NetIOCounter windows interface behavior
2016-10-27 00:44:21 +09:00
5af5f08785
Fix NetIOCounter windows interface behavior
...
addresses a few things:
- Windows has a concept of both a network "interface" and an "adapter"
- These are almost always a one-to-one relationship, though there can be
esoteric instances where they are not.
- I believe the gopsutil NetIOCounters function should only return on a
per-interface level, since this is the behavior on linux/darwin.
Previously, the plugin was basically ignoring the actual interfaces
returned from net.Interfaces(). Instead, it was looping over the net
adapters for each interface, somewhat uselessly.
FWIW, the code for getAdapterList() doesn't exist in the Go standard lib
anymore.
closes #245
2016-10-25 16:51:40 +01:00
6f43e5d707
Merge pull request #274 from leitao/master
...
Improve CPU identification for POWER processors
2016-10-25 10:39:00 +09:00
286927a039
Improve CPU identification for POWER processors
...
Currently gopsutils fails to indentify the POWER processors family,
returning an almost empty Info() structure.
This patch improves the POWER identification without changing what is
available for x86.
2016-10-24 14:11:12 -04:00
c2dd8ca3d4
Merge pull request #269 from andhe/ppc64le
...
Allow clock field in /proc/cpuinfo as cpu MHz fallback value
2016-10-15 21:41:39 +09:00
9d51bfe3bb
Allow clock field in /proc/cpuinfo as cpu MHz fallback value
...
Needed on ppc64le debian porter boxes atleast.
See #230
2016-10-14 14:05:32 +02:00
fc800d3fb4
Merge pull request #267 from botherder/master
...
Added Terminate() for Windows
2016-10-12 00:28:17 +09:00
2facc8e8ed
Merge pull request #265 from hosting-de-labs/add-net-uid
...
add Uids field to net_linux
2016-10-12 00:26:04 +09:00
d44bd00182
Added Terminate() for Windows
2016-10-11 15:19:00 +02:00
e004ef15e1
remove unused code
2016-10-11 10:34:47 +02:00
d6f5a9e920
fix net_test.go
2016-10-06 15:46:14 +02:00
98a0a30dca
update test
2016-10-06 15:38:56 +02:00
ab24c97439
break import cycle
2016-10-06 15:32:25 +02:00
cba0992ab3
add Uids field to net_linux
2016-10-06 14:50:03 +02:00
14eb7acb23
Merge pull request #264 from hillu/percent
...
cpu.Percent() on Windows is very slow.
2016-09-30 21:56:23 +09:00
832dcb96c8
cpu.Percent (Windows): Use the same implementation as on Unix
...
WMI is way too slow.
2016-09-27 18:15:05 +02:00
af2b5127ea
Merge pull request #263 from digitalis-io/master
...
added weightedIO field to disk IOCountersStat (last field of /etc/mtab)
2016-09-22 21:59:23 +09:00
31f4fb8dee
fixed test TestDiskIOCountersStat_String to consider new weightedIO field
2016-09-22 00:03:48 +02:00
acf3fef036
added weightedIO field to disk IOCountersStat (last field of /etc/mtab)
2016-09-21 23:48:49 +02:00
5332238407
Merge pull request #261 from monstermunchkin/issues/260
...
process: make `ClockTicks` arch-independent
2016-09-19 22:31:52 +09:00
68ad8d603c
process: make `ClockTicks` arch-independent
...
The value for `ClockTicks` is defined as `100` by the Linux kernel for
all currently supported architectures in Go. Therefore, there is no need
to define this constant for each architecture separately.
This fixes #260 .
Signed-off-by: Thomas Hipp <thomashipp@gmail.com>
2016-09-17 18:06:07 +02:00
c2c33439b4
Merge pull request #259 from monstermunchkin/issues/258
...
process: determine page sizes via function
2016-09-16 17:42:39 +09:00
eb4a57117f
process: determine page sizes via function
...
Instead of hard-coding the page size for linux systems, use Go's
`Getpagesize` function.
This resolves #258 .
Signed-off-by: Thomas Hipp <thipp@suse.de>
2016-09-16 09:17:34 +02:00
214acac9ea
Merge pull request #256 from shirou/change-platform-version-on-darwin
...
[host]darwin: change to use sw_vers to get platform version
2016-09-13 00:34:47 +09:00
4bab832fc5
fix broken CurrentStatus table.
2016-09-13 00:25:04 +09:00
b9067179ef
[host]darwin: change to use sw_vers to get platform version
2016-09-12 08:07:12 +09:00
f20771d32a
Merge pull request #255 from lberruti/fix_swap_mem_linux
...
Fix: wrong swap values on i686 / 4GB ram.
2016-09-07 14:41:21 +09:00
ea490ab3c5
Merge pull request #254 from bclermont/fix-multidigit-interface
...
OSX Workaround: support interface with multiple digits
2016-09-07 14:25:29 +09:00
3f96312057
only run ifconfig/netstat if necessary, add some tests
2016-09-04 16:37:28 +02:00
36f47562c0
Fix: wrong swap values on i686 / 4GB ram.
...
According to sysinfo manpages: swap fields need to be interpreted along
with the mem_unit (Unit) field.
See also http://stackoverflow.com/a/4229727
2016-09-04 12:39:34 +02:00
145e48efdb
actually there is 13 columns, not 12
2016-08-31 20:41:42 +02:00
57ffe3b1ef
Support interface with multiple digits
2016-08-31 20:41:11 +02:00
859c81da50
Merge pull request #253 from dvusboy/standardize_mhz
...
Standardize `Mhz` to mean maximum CPU frequency on Linux platform
2016-08-30 08:42:49 +09:00
3dc4e52844
Handle the case when the `cpufreq` sysfs does not exist
...
* on virtualized host, this may happen.
* but we may have a value from parsing `/proc/cpuinfo`
* in this case, we do not return the error if we fail to extra
a value from `cpufreq/cpuinfo_max_freq`
2016-08-29 16:21:17 -07:00
10a1ae2123
Standardize `Mhz` to mean maximum CPU frequency on Linux platform
...
* resolve #249
* in `cpu_windows.go`, `Mhz` is the value of `MaxClockSpeed`
* on Linux platform, the `Mhz` value is extracted from `/proc/cpuinfo`
which reflects the current clock speed; treat this as the fallback
value instead
* read from `cpufreq/cpuinfo_max_freq` under sysfs to get the
maximum clock speed for `Mhz`, just like for Windows platform
* also fix the path to `cpu.CoreID` value; the filename is `core_id`
2016-08-29 10:11:30 -07:00
79184fee44
Merge pull request #251 from theothertomelliott/master
...
Correctly handle long running processes on OSX
2016-08-29 14:31:36 +09:00
08a6ab66b0
disk: If could not get Inodes, return empty.
2016-08-27 14:02:51 +09:00
a8e24d70f9
Merge pull request #247 from lfittl/fallback-for-unsupported-architectures
...
Add fallback code for all unsupported operating systems
2016-08-27 00:58:16 +09:00
1435987652
Correctly handle long running processes on OSX
...
The full etime format is dd-hh:mm:ss. Replacing '-' with ':' fixes parsing of this format.
2016-08-26 11:49:57 -04:00
0130d5f088
Merge pull request #248 from dvusboy/cpu_freq_unit
...
Addressing frequency unit discrepancies
2016-08-26 00:06:39 +09:00
123a6c9b0d
Addressing frequency unit discrepancies
...
* for Darwin, it is a minor tweak for readability: the value
returned is in Hz, so using a variable named 'hz' makes more
sense than 'mhz'
* for Linux, the unit is in kHz so we need to divide the value
from `cpuinfo_max_freq` by 10^3 to get MHz (see
cpu-freq/user-guide.txt of the kernel documentation)
2016-08-23 17:21:30 -07:00
3562fb8aa2
Merge pull request #241 from lfittl/add-host-kernel-version
...
Add separate kernelVersion field to host information
2016-08-23 21:11:47 +09:00
58133410ec
Merge pull request #240 from lfittl/fix-osx-linter-warning
...
CPU: Fix linter warning on OSX
2016-08-23 21:08:08 +09:00
41852ca459
Merge pull request #246 from theothertomelliott/master
...
Implement CreateTime for darwin.
2016-08-23 08:03:37 +09:00
Shirou WAKAYAMA