Commit graph

107 commits

Author SHA1 Message Date
Karsten Weiss a8d7d1101a cpu: Support processor-less (memory-only) NUMA nodes (#734)
* cpu: Support processor-less (memory-only) NUMA nodes

Processor-less (memory-only) NUMA nodes exist e.g. in systems that use
Intel Optane drives for RAM expansion using Intel Memory Drive
Technology (IMDT).

IMDT RAM expansion supports two modes:

* "Unify Remote Memory domains": present a processor-less (memory-only)
  NUMA domain, which is the default
* "Expand local memory domains": to expand each processor’s memory domain
  with a portion of the memory made available by Optane and IMDT

This commit fixes a crash in the first case (when "cpulist" is empty).

Here's an example of such a system:

$ numastat -m|head -n5

Per-node system memory usage (in MBs):
                          Node 0          Node 1          Node 2           Total
                 --------------- --------------- --------------- ---------------
MemTotal               118239.56       130816.00       464384.00       713439.56

$ for i in {0..2}; do echo -n "$i: " ; cat /sys/bus/node/devices/node$i/cpulist ; done
0: 0-7,16-23
1: 8-15,24-31
2:

$ /opt/vsmp/bin/vsmpversion -vvv
Memory Drive Technology: 8.2.1455.74 (Sep 28 2017 13:09:59)
System configuration:
    Boards:      3
       1 x Proc. + I/O + Memory
       2 x NVM devices (Intel SSDPED1K375GAQ)
    Processors:  2, Cores: 16, Threads: 32
        Intel(R) Xeon(R) CPU E5-2667 v4 @ 3.20GHz Stepping 01
    Memory (MB): 713472 (of 977450), Cache: 251416, Private: 12562
       1 x 249088MB   [262036/   678/12270]
       1 x 232192MB   [357707/125369/  146]  82:00.0#1
       1 x 232192MB   [357707/125369/  146]  83:00.0#1

* cpu: rename some variables (pkg => node)

* cpu: Use %v not %q in log.Debugf() format strings
2017-11-10 15:31:26 +01:00
Ben Kochie ea250d73f4
Fix off by one in Linux interrupts collector (#721)
* Fix off by one in Linux interrupts collector

* Fix off by one in CPU column handler.
* Add test.

* Enable interrupts in end-to-end test.
2017-11-02 09:59:46 +01:00
Matt Layher f9ad88fc03
xfs: expose correct fields, fix metric names 2017-10-20 18:41:51 -04:00
Ben Kochie deadfef4c9 Update vendoring (#685)
* Update vendor github.com/coreos/go-systemd/dbus@v15

* Update vendor github.com/ema/qdisc

* Update vendor github.com/godbus/dbus

* Update vendor github.com/golang/protobuf/proto

* Update vendor github.com/lufia/iostat

* Update vendor github.com/matttproud/golang_protobuf_extensions/pbutil@v1.0.0

* Update vendor github.com/prometheus/client_golang/...

* Update vendor github.com/prometheus/common/...

* Update vendor github.com/prometheus/procfs/...

* Update vendor github.com/sirupsen/logrus@v1.0.3

Adds vendor golang.org/x/crypto

* Update vendor golang.org/x/net/...

* Update vendor golang.org/x/sys/...

* Update end to end output.
2017-10-05 16:20:47 +02:00
Karsten Weiss b0d5c00832 cpu: Metric 'package_throttles_total' is per package. (#657)
* cpu: Metric 'package_throttles_total' is per package.

'package_throttles_total' is per package, not per cpu. This also reduces
the total number of cpu time series a lot (esp for multi core cpus).

* cpu: Better handling of a cpulist edge-case.

* cpu: Extract the package number from the directory name.

Do not rely on the range index.

* cpu: Add package_throttle_count for node0 cpu1

This file must be ignored by the cpu collector.
2017-09-07 23:24:18 +02:00
Ben Kochie 46c31d8a7e Enable IPVS collector by default (#623)
* Silence error output when no IPVS present.
* Enable by default.
* Update end-to-end fixture.
* Update README.
2017-07-26 15:20:28 +02:00
Andrea De Pasquale 1369763067 Change raid0 status line regexp for mdadm collector (#619) 2017-07-20 17:04:33 +02:00
Aleksey Zhukov 7a914e58f2 Add parsing /proc/net/snmp6 file for netstat-linux (#615)
* Add parsing /proc/net/snmp6 file

* add /proc/net/snmp6 fixture

* fix e2e test

* gofmt

* remove unuser variable

* safe checks

* add tests

* change help format
2017-07-08 20:16:35 +02:00
Matt Layher 6e82fd1c56 Add XFS block mapping and block map B-tree stats (#575) 2017-07-07 07:27:52 +02:00
ideaship 8d90276283 Add bcache collector (#597)
* Add bcache collector for Linux

This collector gathers metrics related to the Linux block cache
(bcache) from sysfs.

* Removed commented out code

* Use project comment style

* Add _sectors to metric name to indicate unit

* Really use project comment style

* Rename bcache.go to bcache_linux.go

* Keep collector namespace clean

Rename:
- metric -> bcacheMetric
- periodStatsToMetrics -> bcachePeriodStatsToMetric

* Shorten slice initialization

* Change label names to backing_device, cache_device

* Remove five minute metrics (keep only total)

* Include units in additional metric names

* Enable bcache collector by default

* Provide metrics in seconds, not nanoseconds

* remove metrics with label "all"

* Add fixtures, update end-to-end for bcache collector

* Move fixtures/sys into tar.gz

This changeset moves the collector/fixtures/sys directory into
collector/fixtures/sys.tar.gz and tweaks the Makefile to unpack the
tarball before tests are run.

The reason for this change is that Windows does not allow colons in a
path (colons are present in some of the bcache fixture files), nor can
it (out of the box) deal with pathnames longer than 260 characters
(which we would be increasingly likely to hit if we tried to replace
colons with longer codes that are guaranteed not the turn up in regular
file names).

* Add ttar: plain text archive, replacement for tar

This changeset adds ttar, a plain text replacement for tar, and uses it
for the sysfs fixture archive. The syntax is loosely based on tar(1).

Using a plain text archive makes it possible to review changes without
downloading and extracting the archive. Also, when working on the repo,
git diff and git log become useful again, allowing a committer to verify
and track changes over time.

The code is written in bash, because bash is available out of the box on
all major flavors of Linux and on macOS. The feature set used is
restricted to bash version 3.2 because that is what Apple is still
shipping.

The programm also works on Windows if bash is installed. Obviously, it
does not solve the Windows limitations (path length limited to 260
characters, no symbolic links) that prompted the move to an archive
format in the first place.
2017-07-07 07:20:18 +02:00
Rene Treffer bcc3cd92b8 Fix cpufreq statistics by converting kHz to Hz 2017-06-27 11:05:55 +02:00
Ben Kochie 182810056f Fix Linux cpu errors (#606)
Make the Linux cpu collector soft-error on missing `cpufreq` and
`thermal_throttle` features.
2017-06-20 07:51:26 +02:00
Rene Treffer 2e9f1913b8 Move stat_linux to cpu_linux and add cpufreq stats (#548) 2017-06-13 11:21:53 +02:00
Emanuele Rocca 047003b6bb Add qdisc collector for Linux (#580)
* Add qdisc collector for Linux

This collector gathers basic queueing discipline metrics via netlink,
similarly to what `tc -s qdisc show` does.

* qdisc collector: nl-specific code moved, names fixed

- netlink-specific parts moved to github.com/ema/qdisc
- avoid using shortened names
- counters renamed into XXX_total

* Get rid of parseMessage error checking leftover

* Add github.com/ema/qdisc to vendored packages

* Update help texts and comments

* Add qdisc collector to README file

* qdisc collector end-to-end testing

* Update qdisc dependency to latest version

Update github.com/ema/qdisc dependency to revision 2c7e72d, which
includes unit testing.

* qdisc collector: rename "iface" label into "device"
2017-05-23 11:55:50 +02:00
Robert Clark 58f50b31f2 Multiply port data XMIT/RCV metrics by 4 (#579)
According to Mellanox, it is standard practice that the port_xmit_data and port_rcv_data
files are split into 4 lanes. To get the actual transmit and receive values for each
port, the metric needs to be multiplied by 4.

Signed-Off-By: Robert Clark <robert.d.clark@hpe.com>
2017-05-12 07:28:53 +02:00
Matt Layher 1feb091b36
Initial XFS collector 2017-04-22 11:53:07 -04:00
Karsten Weiss d9703ff7c6 edac: Fix typo in csrow label of node_edac_csrow_uncorrectable_errors_total metric. 2017-04-18 12:45:06 +02:00
Karsten Weiss 45ca8db352 Support the 'guest_nice' cpu mode of /proc/stat.
'guest_nice' is available since Linux 2.6.33.
2017-04-14 12:50:37 +02:00
Sam Kottler 6eafa51fa8 Add ARP collector for Linux (#540)
* Implement commonalities and linux support for ARP collection

* Add ARP collector to fixtures and run as part of e2e tests

* Bubble up scanner errors

* Use single return values where it makes sense

* Add missing annotation

* Move arp_common into arp_linux

* Add license header to arp_linux.go

* Address initial feedback

* Use strings.Fields instead of strings.Split

* Deal with scanner.Err() rather than throwing away errors

* Check for scan errors in-line before interacting with the entries map

* Don't interact with potentially empty text from scan

* Check for scan errors outside the scan loop

* Add comment about moving procfs parsing

* Add more direct comment

* Update initialism style to match go style guide

* Put function args on the same line

* Add TODO in front of comment about procfs extraction

* Guard against strings.Fields returning an empty slice

* Be more defensive about ARP table format and use upcase more broadly

* Enable the ARP collector by default

* Add ARP collector to the README

* Remove 'entry'
2017-04-11 17:45:19 +02:00
Johannes 'fish' Ziemke 9676f5f2dc Merge pull request #523 from roclark/support-legacy-infiniband
Add support for legacy InfiniBand drivers
2017-03-21 10:52:07 +01:00
Matt Layher 2bfe410fb7
Expand wifi collector for more interface types 2017-03-20 12:25:01 -04:00
Robert Clark 3a5917dfdc Add support for legacy InfiniBand drivers
Older versions of the OFED drivers contain 64-bit variants of the port counters and are located in a directory named 'counters_ext'. This patch includes these older metrics that have since been deprecated with OFED 4.0.

Signed-Off-By: Robert Clark <robert.d.clark@hpe.com>
2017-03-20 10:37:21 -05:00
Tobias Schmidt 0400e437be Fix and simplify parsing of raid metrics
Fixes the wrong reporting of active+total disk metrics for inactive
raids. Also simplifies the code and removes a couple of redundant
comments.
2017-03-19 08:03:58 -03:00
Matt Layher 69368b7f9c Add synthetic node_wifi_station_info metric for BSS information 2017-03-16 16:24:23 -04:00
Brian Brazil a02e469b07 Report collector success/failure and duration per scrape. (#516)
This is in line with best practices, and also saves us
63 timeseries on a default Linux setup.
2017-03-16 17:21:00 +00:00
Tobias Schmidt ce117d7a40 Update vendored packages 2017-02-28 18:20:24 -04:00
Tobias Schmidt d1dfda86ee Fix wrong end-to-end expectation 2017-02-28 16:02:43 -04:00
Ben Kochie 38cd07ebb9 Merge pull request #450 from roclark/add-infiniband
infiniband: Add new collector for InfiniBand statistics
2017-02-16 14:33:19 +01:00
Ben Kochie a097dd36b3 Merge pull request #459 from joehandzik/wip-zpool-io-cherrypick
ZFS Collector: Add zpool IO statistics
2017-02-16 08:16:55 +01:00
Thorhallur Sverrisson 5ab285e098 Adding buddyinfo to end to end test. 2017-02-15 10:15:44 -06:00
Thorhallur Sverrisson 3ba15c1ddb Adding support for /proc/buddyinfo for linux free memory fragmentation.
/prod/buddyinfo returns data on the free blocks fragments available
for use from the kernel.  This data is useful when diagnosing
possible memory fragmentation.

More info can be found in:
* https://lwn.net/Articles/7868/
* https://andorian.blogspot.com/2014/03/making-sense-of-procbuddyinfo.html
2017-02-15 10:15:43 -06:00
Joe Handzik bb8b3fca88 ZFS Collector: Add zpool IO statistics
Signed-Off-By: Joe Handzik <joseph.t.handzik@hpe.com>
2017-02-10 13:31:25 -06:00
Robert Clark 4866adcb71 Add new collector for InfiniBand statistics
Add new metrics for the InfiniBand network protocol including the amount of packets sent and received, the number of times the link has been downed and how many times the link has recovered from an error state.

Signed-Off-By: Robert Clark <robert.d.clark@hpe.com>
2017-02-07 11:09:08 -06:00
Joe Handzik 8c23f5ff54 ZFS Collector: Convert dashes to underscores for metrics
This fixes #442, and prevents other ZFS metrics from slipping through in the future.

Signed-Off-By: Joe Handzik <joseph.t.handzik@hpe.com>
2017-01-31 14:11:56 -06:00
Ben Kochie 71362d45eb Merge pull request #432 from joehandzik/wip-zfs-zfetchstats
Update ZFS Collector with most non-zpool metrics
2017-01-31 08:52:41 -05:00
Joe Handzik e5ee274a32 ZFS Collector: Move from camelcase to underscores for metric prefixes
Signed-Off-By: Joe Handzik <joseph.t.handzik@hpe.com>
2017-01-29 15:59:01 -06:00
Ben Kochie 5a6db5c8d2 Handle multiple NFS device mounts
It's possible to mount an NFS share in multiple locations.
* Duplicates contain the same metric values, so they can be ignored.
* Update fixture.
2017-01-24 13:44:08 +01:00
Joe Handzik 94fb93a9f3 ZFS Collector: Add dmu_tx functionality
Signed-Off-By: Joe Handzik <joseph.t.handzik@hpe.com>
2017-01-23 16:41:15 -06:00
Joe Handzik 07c7ae733a ZFS Collector: Add fm functionality
Signed-Off-By: Joe Handzik <joseph.t.handzik@hpe.com>
2017-01-23 16:31:22 -06:00
Joe Handzik 05048c067d ZFS Collector: Add xuio_stats functionality
Signed-Off-By: Joe Handzik <joseph.t.handzik@hpe.com>
2017-01-23 16:30:37 -06:00
Joe Handzik 3c9e779989 ZFS Collector: Add vdev_cache_stats functionality
Signed-Off-By: Joe Handzik <joseph.t.handzik@hpe.com>
2017-01-23 16:29:50 -06:00
Joe Handzik a02ca9502c ZFS Collector: Add zil functionality
Signed-Off-By: Joe Handzik <joseph.t.handzik@hpe.com>
2017-01-23 16:29:00 -06:00
Joe Handzik a3125ab4d9 ZFS Collector: Add zfetchstats functionality
Signed-Off-By: Joe Handzik <joseph.t.handzik@hpe.com>
2017-01-23 16:28:11 -06:00
Johannes 'fish' Ziemke 2884181cce Merge pull request #415 from mdlayher/mountstats-nfs-additional
Add NFS event metrics to mountstats collector
2017-01-12 14:08:21 +01:00
Matt Layher e3f99e13b9
Add NFS event metrics to mountstats collector 2017-01-11 11:41:13 -05:00
Matt Layher efa25665ec
Add initial wifi collector, bump netlink to fix 32-bit builds 2017-01-11 10:08:44 -05:00
Johannes 'fish' Ziemke 55170e8feb Merge pull request #411 from discordianfish/hwmon-move-label-metrics
Use filename as label, move 'label' to own metric
2017-01-10 12:21:18 +01:00
Ben Kochie 38a4a36061 Update end-to-end test. 2017-01-10 10:23:16 +01:00
Ben Kochie b4fa10ca9d Add collector for Linux EDAC
Collect "Error detection and correction" metrics from memory
controllers.
* Supported on Linux only.
* Add basic fixtures.
* Enabled by default.
2017-01-10 10:14:19 +01:00
Johannes 'fish' Ziemke 6aef20f8d8 Use filename as label, move 'label' to own metric
This closes #406
2017-01-09 18:33:31 +01:00