Copy stable/10@r272459 to releng/10.1 as part of
the 10.1-RELEASE process.

Approved by: re (implicit)
Sponsored by: The FreeBSD Foundation

Copy head (r256279) to stable/10 as part of the 10.0-RELEASE cycle.

Approved by: re (implicit)
Sponsored by: The FreeBSD Foundation

hrStorageSize and hrStorageUsed are 32 bit integers, reporting a fs
size and usage in hrStorageAllocationUnits. If the file system has
more than 2^31 allocations it can not be shown correctly and the
meters are useless.

In such cases follow net-snmp behaviour and increase
hrStorageAllocationUnits so the values fit under INT_MAX.

PR: bin/177183
Submitted by: Eugene Grosbein egrosbein rdtc.ru
MFC after: 2 weeks

In snmp_hostres, device_map table is used for consistent device table
indexing. When a device has gone it is not removed from device_map
table but just its entry_p field is set to NULL.

So when traversing device_map in disk_OS_get_ATA_disks() and
disk_OS_get_MD_disks() check for entry_p being NULL, otherwise the
bsnmpd crash is possible when a removed map entry is dereferenced.

Before the fix, for disk_OS_get_ATA_disks() the crash could be easily
reproduced running:

atacontrol detach ata1

The crash was not observed in disk_OS_get_MD_disks() because currently
snmp_hostres does no see md(4) disks: to get the device list it uses
devinfo(3), which does not return md devices.

Reported by: Miroslav Lachman 000.fbsd quip.cz
MFC after: 1 week

Do not count kernel threads as processes for hrSystemProcesses OID.

PR: bin/160494
Submitted by: Jeremy Chadwick <freebsd jdc.parodius.com>

Spelling.

Spelling fixes for usr.sbin/

Apparently, "ada" drives are better treated similarly to "da" drives.

It seems that the warning is much less severe than its message says. The
device is certainly added to the list after the first pass.

Fix typo

MFC after: 1 month

Use full buffer size in read(2) call, there is no need to preserve the
last byte of the buffer.

Since we call refresh_device_tbl() for any devctl event types - no need
to check the first byte of buffer. Remove these checks.

Also remove logging for the case of unknown devd message. It incorrectly
triggers when all devctl events are not fit into one buffer and part of
unread data will be read in the next pass.

When length of data readed from devctl is equal to sizeof(buf), then try
to read from socket again, to read full data.

MFC after: 2 weeks

[mdoc] Fixed .Dt call.

Fix CPU load reporting independent of scheduler used.

- Sample CPU usage data from kern.cp_times, this makes for a far more
accurate and scheduler independent algorithm.
- Rip out the process list scraping that is no longer required.
- Don't update CPU usage sampling on every request, but every 15s
instead. This makes it impossible for an attacker to hide the CPU load
by triggering 4 samplings in short succession when the system is idle.
- After reaching the steady-state, the system will always report the
average CPU load of the last 60 sampled seconds.
- Untangling of call graph.

PR: kern/130222
Tested by: Julian Dunn <jdunn@aquezada.com>
Gustau Pérez <gperez@entel.upc.edu>
Jürgen Weiß <weiss@uni-mainz.de>
MFC after: 2 weeks

I'm unsure if some MIB standard states this must be the load average
for, eg. 300s, it looks like net-snmp isn't even bothering to implement
the CPU load reporting at all.

Make failed open of /dev/mdctl in the bsnmpd hostres module non-fatal.
This makes it possible to use the hostres module when bsnmpd is not
running as root.

MFC after: 1 week

mdoc: consistently spell our email addresses <foo@FreeBSD.org>

Reviewed by: ru

Port all applications in usr.sbin/ from libulog to utmpx.

Let the snmp_hostres module use utmpx.

Approved by: harti

Collapse devinfo_state_t with device_state_t in order to avoid a
structure replication and improve manteneability.

Reviewed by: jhb, imp
Tested by: Riccardo Torrini <riccardo at torrini dot org>

add zfs oid to bsnmpd

PR: bin/129360
Submitted by: Ulrich Spoerlein

! is a generic info message

Reviewed by: harti
Approved by: harti
MFC after: 2 weeks

Revise markup.

Reduce the memory requirements for the tables by (1) using malloced strings
instead of maximum sized arrays embedded in the table structures and (2)
using pointers to constant oids instead of copying the oid into each
table structure. This also fixes indexing in the case when a string used
for indexing is longer than the maximum size allowed in the MIB.

Submitted by: Victor Cruceru <soc-victor@>

Expand contractions.

Replace ki_ocomm by ki_comm. The former will go away.

Submitted by: Victor Cruceru <soc-victor>
Suggested by: Julian Elischer

Fix a bug in the processor load code: the wrong process was used to
get the inital processor load. Also fix compilation on sparc64 without NDEBUG.

Fix false positives when searching for disks. Things like adv(4) are not
disks.

Submitted by: Victor Cruceru <soc-victor@>

Fix the MIB to use libgeom instead of libdisk.

Submitted by: Victor Cruceru <soc-victor>

Fix build without -DNDEBUG.

Spotted by: obrien

assert(3) is not used here.

msg