| |
VK2GDM > BPQ 06.02.96 02:39l 201 Lines 8002 Bytes #-11230 (0) @ WW
BID : 49226_VK2GDM
Read: DB3RP DH9MEC DL1AMQ DG0EL GUEST
Subj: ## Commands No2 of 2 ##
Path: DB0AAB<DB0SL<DB0RGB<DB0ABH<DB0BOX<DB0LPZ<DB0MER<DB0BAL<DB0JES<DB0GR<
DB0BLO<DB0TEM<DB0GWD<DB0HRO<DB0OQ<DB0HES<OZ7BOX<OZ2BOF<OZ9PAC<OZ4PAC<
OZ4BBS<OZ3PAC<OZ6BBS<SK7CY<SK7BK<SM7DLZ<N0RSE<KT0H<N0QCU<WX3K<KA6EYH<
VK2TLH<VK2BXQ<VK2GDM
Sent: 960126/0313Z @:VK2GDM.#SYD.NSW.AUS.OC #:49226 [Airds] FBB5.15c $:49226_VK
From: VK2GDM@VK2GDM.#SYD.NSW.AUS.OC
To : BPQ@WW
by the software - it is just information for the user. The '!' indicates
a 'locked route' - one entered in the CONFIG file, or via SYSOPH. Other
entries come and go as this node hears NODES broadcasts.
Additional information may be obtained by entering r r
r r
NOTTS:G8BPQ-3} Routes:
4 GB7YS 0 0! 8 0 0% 0 0 18:52
1 G4FIS-2 14 0! 0 0 * 0 0 00:00
3 G3SDC-8 35 0! 0 0 * 0 0 00:00
> 6 G4FIS-2 150 58! 57022 1333 2% 0 0 21:55 0
3 G0INA-1 90 0! 0 0 * 0 0 00:00
2 G8BPQ-5 250 0! 0 0 * 0 0 00:00
4 G1EQT-8 50 0! 0 0 * 0 0 17:13
> 4 G0INA-3 120 48! 19305 819 4% 0 0 21:40 0
> 5 GB7YS 150 93! 27017 6229 23% 0 0 21:44 0
3 G7JGX-3 25 0! 0 0 * 0 0 00:00
> 4 G4IRX-3 50 9! 2025 42 2% 0 0 22:09 0
7 G8BPQ-8 250 105 592 0 0% 0 0 22:02
4 G7EQM-8 10 1 3077 219 7% 0 0 21:37
The extra fields are:
Info frames sent
Info frames retransmitted
Retry Rate - the ratio of the above 2, as a percentage
(or * if both are zero)
Non-standard maxframe (from Config file)
Non-standard frack ( ditto )
Time the last NODES broadcast was heard from this node.
The last figure is only present if there is an active link. It
is the number of frames queued to be sent. Up to 4.05e, this only
counts frames queued above the link level - there may be up to
another 8 queued at link level. With version 4.05f and above it
includes all frames queued.
The retry rate gives a good indication of how well the link is running.
Dedicated links should normally be well below 10%. A shared link
will normally have a higher rate, but anything above say 25% is likely
to cause significant delays. Another indication of a poor link is a
high number of frames queued - any nonzero value of 4.05e or below, or
above 8 for later versions which persists for more that a minute or so
is likely to be causing problems.
The USERS command lists the stations currently using the node.
NOTTS:G8BPQ-3} G8BPQ Network System V3.21 (95)
Host6(NOTTS:G8BPQ-3)
Host3(NOTTS:G8BPQ-3) <--> Circuit(LRG7:G0GDR-1 G8BPQ-1)
Uplink(G9XXX) <~~> Downlink(G9YYY)
Host is an internal (Normally BBS Port)
Circuit is a link from/to another node.
Uplink is a connection from a normal Ax.25 station.
Downlink is a connection to a normal user.
The <--> indicates an active session. <~~> indicates a session being set up.
The Number on the end of the header line is the number of free buffers.
There are a few commands not given in the menu. These are primarily of
interest to the sysop, (or for me to experiment with). They are LINKS,
STATS, L4T1, and PACLEN.
The LINKS command lists the currently active AX.25 Sessions (Both user
access and node-node links)
NOTTS:G8BPQ-3} Links:
G0GDR-1 G8BPQ-3 S=5 P=2 T=3 V=2
S is the link state (see AX25 protocol spec, but the main ones are
2 (connecting) 4 (disconnecting) 5 (connected)).
P is the port.
T is the link type. 1=Uplink, 2=Downlink, 3=Node-Node link.
V is the AX.25 Version (1 or 2).
The STATS command displays a number of counters.
s
NOTTS:G8BPQ-3}
Time active (mins) 8786
Timer Overruns 48
Buffers: Max/Cur/Min/Out 100 76 16 0
Known Nodes 144
L4 Connects Sent/Rxed 1055 1732
L4 Frames TX/RX/Resent/Reseq 73288 96544 1870 42
L3 Frames Relayed 10148
Port 1 Port 2 Port 3 Port 4 Port 5 Port 6 Port 7
L2 Frames Digi-ed 0 0 0 0 0 0 0
L2 Frames Heard 124393 1870 80601 82124 60600 167575 2650
L2 Frames Rxed 11921 0 2909 66674 59032 94340 1447
L2 Frames Sent 15955 1870 4690 69857 71328 107267 3284
L2 Timeouts 4466 0 984 5437 7697 1154 58
REJ Frames Rxed 7 0 2 460 1681 214 0
RX out of Seq 269 0 3 951 1697 272 0
Undrun/Poll T-O 0 0 5358 233 0 0 0
RX Overruns 0 0 39814 0 0 0 63
RX CRC Errors 42260 0 1 0 2324 27025 0
FRMRs Sent 0 0 0 19 0 0 0
FRMRs Received 0 0 0 0 0 0 0
Frames abandoned 1 0 0 0 0 2 0
Link Active % 9 64 0 0 0 0 0 0 14 21 50 82 0 0
If you have more than 7 ports the display starts to look a bit untidy, so
there are now two options on the STATS command:
S S Displays just the system and L3/4 stats
S n Displays just the port stats, starting with port N,
for a maximum of 7 ports. So if you have 10 ports,
S 1 displays the first 7, and S 8 displays 8,9 and 10.
Most are fairly obvious, but a few need a bit of explanation. There is one
set of level 2 counters for each port.
Time Active is the time since the system was loaded.
The Buffer count Maximum, Minimum and Current should be obvious. The
Out figure is the number of times a request to allocate a buffer
failed because there were none available.
The L4 frames resent count is the number retransmitted because an
ACK wasnt received within the L4 timeout period. The Reseq count
is the number of frames received out of sequence, but subsequently
used because the missing frame(s) eventually arrived.
Timer Overruns count the number of times the main code (which is initiated
by the timer interrupt) is still running when the next timer interrupt
occured. If it increments very rapidly (several times a minute), it
indicates that the PC is too slow for the amount of data being processed.
It is normal for an XT class machine running a couple of BBS ports to get
about one overrun per minute. My 10MHz AT clone shows very few.
A large number in the 'REJ received' field may indicate that your Maxframe
is too high. Similarly a large number in 'RX out of sequence' may indicate
the the station taking to you has too large a maxframe.
RX Overruns indicate characters lost because the software didn't process
the interrupts fast enough. If you are running KISS ports, and you get
a lot (ie a significant percentage of L2 frames heard), try reducing the
speed of the link from the PC to the TNC. If you are using an HDLC card
(DRSI or PC120), particularly at high speed, then there isnt much you
can do except buy a faster machine. (But I would like to hear from anyone
having problems running at 9600 baud or above - I may be able to speed up
the routines a bit).
Underuns indicate a similar problem in responding to TX interrupts,
but only apply to HDLC cards. As you cant get a TX underrun on an
async port, this field is also used to count timeouts on a polled KISS
system.
Frames abandoned counts the number of frames discarded because they have
been waiting to be sent (for DCD to clear) for more than 60 secs. If a
lot occur, then either your squelch is a bit dodgy, or the channel is
VERY busy. Only used for HDLC cards.
Link Active shows two values for each port. The first is the % of time
in the last minute that your station was transmitting, and the second
the % the channel was active (sum of Transmitting and DCD active). Only
maintained for HDLC ports. Note that of you are using SOFTDCD, then the
indicated % active may be an overestimate.
The L4T1 command displays or sets the Level 4 timeout used for the current
session. It is primarily for me to experiment with.
PACLEN sets or displays the PACLEN value used for messages generated by
the node (eg command responses). Again it is mainly for me to play with.
John Wiseman, G8BPQ
Revised 24/5/93
Read previous mail | Read next mail
| |
