Project

General

Profile

Configuring the ipaccess nano3G » History » Version 33

neels, 12/14/2018 02:06 PM

1 2 neels
{{>toc}}
2 1 neels
3 2 neels
h1. Configuring the ip.access nano3G
4
5
h2. IP address
6
7 21 neels
The ip.access nano3G will obtain an IP address from the DHCP server in your network. Ways to find out:
8
9
* Run @abisip-find@ (currently in the osmo-bsc.git repository), which should find the nano3G (despite no Abis interface being present). Find the @abisip-find@ output matching the MAC address printed on the sticker next to the power jack.
10
* Watch wireshark, filtering on 'BOOTP' while the nano3G starts up.
11
* Look it up in your DHCP server's list of leases.
12
13 2 neels
For this text, let's assume the IP address it obtained is 192.168.0.124.
14
15
h2. Initial Config
16
17
Once off, configure:
18
19
* the MCC + MNC,
20
* the UARFCN (i.e. the frequencies to transceive on) and
21 25 tsaitgaist
* the LAC and RAC (max. 256)
22
* the Cell ID composed by RNC (12 bits) and HNB C-ID (16 bits)
23 2 neels
24
You can do this on the _dmi_ console reachable by telnet:
25
26
<pre>
27
telnet 192.168.0.124 8090
28
dmi>
29
</pre>
30
31
On the dmi, enter commands like these:
32
33
<pre>
34
# PLMN Id == MCC + MNC
35
set mcc="901"
36 30 daniel
set mnc="70"
37 2 neels
38
# [uarfcnDownlink, 1900 MHz band], [scramblingCode], [dummyCellId]
39
set rfParamsCandidateList=({9800, 401, 1})
40
41
# [lac], [rac]
42 1 neels
set lacRacCandidateList=({10422, (99)})
43 25 tsaitgaist
44
# cellid == RNCID<<16+HNBCID
45
set hnbCId=1
46
set rncIdentity=0
47
48 2 neels
</pre>
49
50
These settings persist across nano3G power down.
51
52 22 neels
*All of the above settings require a reboot of the nano3G to take effect.*
53
54 23 neels
h2. After Every Boot: Starting Operation
55 2 neels
56
Every time you boot the nano3G, you need to
57
58
* set the IP address the nano3G will find the HNB-GW at.
59
* 2061 = set cell parameters
60
* 1216 = unlock ap
61
* activate HNB-GW connection
62
* set csg to open access so that any IMSI can register
63
64
Enter the _dmi_...
65
<pre>
66
telnet 192.168.0.124 8090
67
dmi>
68
</pre>
69
70
...and issue commands like:
71
72
<pre>
73
set hnbGwAddress="192.168.0.132" 
74
action 2061
75
action 1216
76
action establishPermanentHnbGwConnection
77
set csgAccessMode=CSG_ACCESS_MODE_OPEN_ACCESS
78
</pre>
79
80 29 daniel
It is also possible to change the downlink power for the nano3G. This value is reset after the nano3G reboots and after changing the value the connection to the hnbgw needs to be reestablished before the change will become active.
81
82
The dmi attribute needed to set is maximumTotalWidebandTransmitPower
83
84
<pre>
85
dmi> set maximumTotalWidebandTransmitPower = 0
86
</pre>
87
88
The default value is 70 and a change of +/-10 seems to affect the output power by about 1dBm.
89
It is not possible to set the attribute to a negative value and setting it to something larger than the default should be done carefully
90
91
In our tests we were able to measure a tx power output of -6.5dBm for 1960MHz/UARFCN9800 with the attribute set to the lowest value (0) and 0dBm, with the default value of 70.
92
93
94 2 neels
h2. SSH Access
95
96
The nano3G come with a root password of 'newsys':
97
<pre>
98
ssh root@192.168.0.124
99
password: newsys
100
</pre>
101
102 9 roox
In case you are using a recent version of the OpenSSH-client you'll get the following error message while trying to connect:
103
_Unable to negotiate with 192.168.0.124 port 22: no matching key exchange method found. Their offer: diffie-hellman-group1-sha1_
104
105
Starting with OpenSSH version 7.0 support for the 1024-bit diffie-hellman-group1-sha1 key exchange was disabled by default at run-time.
106 24 neels
Furthermore, the ciphers supported by the nano3G may be disabled by default. To enable both manually:
107 9 roox
<pre>
108 24 neels
ssh -o KexAlgorithms=+diffie-hellman-group1-sha1 -c aes128-cbc root@192.168.0.124
109 9 roox
</pre>
110
111 10 roox
h2. NTP
112
113
Be sure that the nano3G is able to resolve the DNS record _0.ipaccess.pool.ntp.org_ and can connect the the corresponding NTP servers.
114
Without syncronized NTP the nano3G does not bring up the TRX and it even do not try to connect to the hnbGw.
115 9 roox
116 20 neels
A workaround is to edit the @/etc/hosts@ file on the nano3G (via SSH) and add an entry to resolve 0.ipaccess.pool.ntp.org to your local NTP server's IP address (which requires re-mounting the root file system in read-write mode...).
117 19 neels
118 2 neels
h2. Logging
119
120
When logged in via SSH, you can view the live logging here:
121
<pre>
122
ls /tmp/iapclogs/trace_*.log
123
</pre>
124
125
h2. Closed Mode
126
127
You can also set csgAccessMode to closed and allow only specific IMSIs:
128
129
<pre>
130 6 neels
set csgAccessMode=CSG_ACCESS_MODE_CLOSED_ACCESS
131 2 neels
# IMSI, 1:allowed/2:not allowed, phone number (only for "Closed Access")
132
set accessControlList = ({"001010123456015",  1, "81084"},{"001010123456025", 2, "81025"})
133
</pre>
134
135
(The phone number is actually not relevant)
136
137 27 tsaitgaist
h2. Neighbour Cells
138
139
If you have multiple cells running you can add the other to the neighbouring cell list over dmi.
140
141
<pre>
142
set neighbourListPopulation=STATIC_ONLY
143
# ({band:ENUM, dlUarfcn:INT, scramblingCode:INT, mcc:STRING, mnc:STRING, lac:INT, rac:INT, rncId:INT, cellId:INT, qQualMin:INT, qRxlevMin:INT, txDiversityIndicator:ENUM, primaryCpichTxPower:INT, hcsPriority:INT, qHcs:INT, deltaQhcs:INT}, ...)
144
set staticUmtsNeighbourList_001=({UMTS_BAND_2, 9800, 401, "262", "42", 10422, 99, 0, 1, -24, -119, TX_DIVERSITY_DISABLED, -10, 0, 0, -5},)
145
</pre>
146
147 2 neels
h1. Peculiarities and Tips
148
149 12 neels
h2. Exiting the dmi while keeping it alive
150
151
When you exit the dmi telnet by hitting Ctrl-C, it will not be available anymore until you reboot the nano3G.
152
Every connection attempt will then end in
153
154
<pre>
155
dmi> Connection closed by foreign host.
156
</pre>
157
158 1 neels
However, if you end your session by the telnet escape character and quitting, the dmi remains open for further connections. Usually that means: hit Ctrl-] (Ctrl and closing square brace) and then enter 'quit':
159
160
<pre>
161
dmi> ^]
162
telnet> quit
163 28 tsaitgaist
</pre>
164
165
If the telnet interface for dmi is closed (after you quit it), you can still issue commands using the [[Configuring_the_ipaccess_nano3G#SSH-Access|console]] (here to close the connection to the HNB-GW)
166
<pre>
167
/opt/ipaccess/DMI/ipa-dmi -c 'action terminatePermanentHnbGwConnection'
168 12 neels
</pre>
169
170 2 neels
h2. UE Register
171
172
The nano3G apparently passes the same identity received from the UE through to
173
the HNBAP UE Register Request message. This means that when the UE sends a
174
TMSI, the UE Register Request received by osmo-hnbgw contains no IMSI.
175
176 7 neels
In this scenario, the problem is that Paging apparently does not (always) work.
177 1 neels
So even though we have working code that allows HNBAP registration with
178 7 neels
a TMSI, that means that you can't (always) reach the UE from the CN.
179
This is not always the case, sometimes the nano3G can well page UEs that have
180
registered by TMSI. Vague idea: it may be that it needs to have seen the IMSI
181
once after power-cycling, e.g. after a closed-mode registration, and then
182
TMSI registration will not harm Paging. (TODO: clarify this)
183 2 neels
184 7 neels
The VTY configuration option to allow TMSI-only attaching to HNBGW, which
185
possibly helps to shorten your dev cycle but may harm paging, is:
186 2 neels
187
<pre>
188
hnbgw
189
 iuh
190 7 neels
  hnbap-allow-tmsi 1
191 2 neels
</pre>
192 1 neels
193 2 neels
Legacy workaround: connect the phone to a different network between retries (being
194
rejected suffices). That causes the UE to discard its TMSI and then use the IMSI
195
for the next registration.
196
197 18 neels
A [[Configuring the ipaccess nano3G#Closed-Mode|closed csgAccess]] with explicit IMSIs could help here to enforce that a UE
198 7 neels
indeed sends its IMSI to the nano3G and hence Paging should work.
199 3 neels
See also #1924.
200
201
h2. id-Reset
202
203
The nano3G seems to *not* send an id-Reset message upon connecting to the HNB-GW.
204
205
h2. Location Update failure due to timeout
206
207
If a UE seems to connect successfully at first but fails by timeout because the final
208
"TMSI Reallocation Complete" message is missing, this might be due to misconfiguration:
209
the CN is sending the wrong LAC or the PLMN-ID (NCC/MNC) is configured wrongly.
210
211
This might be confusing in the sense that a complete LU worked once but not after that;
212
GMM Attach may be successful; Security Mode Commands succeed; and so forth.
213 8 neels
Still the solution might be simply to fix the mobile network code in the osmo-msc.cfg.
214 4 neels
215
h2. RAB Assignment needs IuUP ACK Initialization
216
217
IuCS uses UP encapsulated in RTP. The UP starts off by sending an Initialization, replied
218
upon by an ACK Initialization.
219
220
The nano3G seems to not reply with an ACK when it receives an IuUP Initialization frame.
221
Thus it is not possible to merely echo its own RTP packets back to itself; instead, the
222
first RTP frame received from the nano3G (that is an IuUP Initialization) can be changed
223
to an ACK Initialization by writing 0xe4 to the first payload byte. Sending this back to
224
the nano3G then results in successful RAB Assignment.
225
226
(With the SysmoCell5000, echoing its own Initialization back to itself results in an ACK
227 1 neels
being sent, which we can also echo back to itself, so mere echoing works there.)
228 11 neels
229 14 neels
A hack to make the nano3G work is currently on the "3G master" (the vlr_3G branch),
230
because it does not seem to harm other femto cells: the commit's summary is
231
"mgcp: hack RAB success from nano3G: patch first RTP payload", see "cgit":http://git.osmocom.org/openbsc/log/?h=vlr_3G.
232 15 neels
233
See also #1712#note-21 and the following two comments.
234 31 lynxis
235
h2. There is no DMI interface and the device restarts every 5 minutes
236
237
The ssh should still work. One of many daemons might not been started. This happens usually when a daemon is using a .cfg which is meant for another version.
238
Removing the configuration helps, because every daemon have a default configuration.
239
240 32 lynxis
**cat /tmp/consoleLog** shows what happened! <pre>grep FAILURE /tmp/consoleLog</pre>
241 31 lynxis
242
The configuration files are located under **/var/ipaccess/config_bank_2**
243 33 neels
244
245
h2. Phone is registered, but is usually not receiving any SMS / voice calls
246
247
In open access mode, it seems the nano3G sometimes fails to page a phone -- it looks like it didn't store the subscriber's IMSI. Try to register the phone that should receive a call to a different network (being rejected suffices, just to get the TMSI out of the phone's state) and then re-register to the hNodeB.
248
249
It may also help to set the nano3G to closed access and indicate your IMSI as one of the allowed subscribers.
Add picture from clipboard (Maximum size: 48.8 MB)