From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============0071511886316583283==" MIME-Version: 1.0 From: Jonas Bonn Subject: Re: Unsalble cellular connection Date: Tue, 13 Aug 2019 17:00:25 +0200 Message-ID: In-Reply-To: List-Id: To: ofono@ofono.org --===============0071511886316583283== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable CC: ofono mailing list. Please don't drop the mailing list from the conversation. On 13/08/2019 13:51, JH wrote: > Hi Jonas, > = > On 8/13/19, Jonas Bonn wrote: >> Hi JH, >>> Aug 01 08:09:39 connmand[181]: wwan0 {add} route 0.0.0.0 gw >>> 10.114.23.146 scope 0 >>> Aug 01 08:09:45 connmand[181]: Online check failed for 0xa71cf8 Telstra > = >> Online check failed... was the network connection ever working? > = > Yes, it usually in good LTE connection when power up, then it dropped > connection in about 1 - 2 hours. > = >>> Aug 01 09:34:11 connmand[181]: Time request for server 10.114.23.146 >>> failed (101/Network is unreachable) > = >> Is this attempt to check the time server causing the modem to discover t= hat the connection >> isn't working? Who sets the network interface DOWN here... the QMI Linu= x driver? Why >> doesn't the modem signal something on the QMI communication channel wh= en this >> happens... maybe it does but the message is just being silently ignored? > = > $ ping 10.114.23.146 > --- 10.114.23.146 ping statistics --- > 5 packets transmitted, 0 packets received, 100% packet loss > = > Cannot reache that IP address?? That address is the gateway address for that link... there's unlikely a = timeserver running there. Why does connman think there's a timeserver = there? > = >> There's very little traffic on the link... is the PDP context silently e= xpiring due to inactivity? > > What happens if you ping something every minute or so when the link > is active? > = > That because all traffic went to WiFi, the application is sending a > message to Cloud every minitue, should that be better than to ping? So you have no traffic over the LTE link? How do you know that it goes = down after 1-2 hours... maybe it went down after 5 minutes? Cloud messages are as good as a ping... a packet's a packet. > = >>> Aug 01 09:34:11 connmand[181]: wwan0 {del} address 10.114.23.145/30 lab= el wwan0 >>> Aug 01 09:34:11 connmand[181]: Skipping disconnect of >>> /ubloxqmi_0/context1, network is connecting. > = >> Here I think ofono and connman just get out of sync because connman seem= s to know that >> the network is down while ofono doesn't... > = > Does that alude it is not the ofono caused link down? The link is configured by connman, not ofono. ofono just publishes the = address details so that connman can set it up. > = >> Sure... there's not much to go on in the logs you've posted, just an >> assertion that the LTE connection was silently dropped. It would be >> better if you could show the behaviour with more complete debug output: >> >> ofonod -d >> >> The issue could be (perhaps) that there's a QMI message that ofono >> doesn't handle and is silently ignoring... dumping the QMI communication >> might useful: >> >> OFONO_QMI_DEBUG=3D1 ofonod -d > = > Before running following commands, the LTE connection dropped, after > running following commands, LTE connection came back, it is wired that > the LTE is more stable than running by system service where > ExecStart=3D/usr/sbin/ofonod -u", LTE connection has not been dropped > for more than 6 hours, I'll run it over night and let you know if the > connection will be dropped or not. Why running OFONO_QMI_DEBUG=3D1 > ofonod -d makes it more stable? Can I say that LTE drop off is caused > by the ofono? > = > # systemctl stop ofono > # OFONO_QMI_DEBUG=3D1 ofonod -d > # ip addr > 1: lo: mtu 65536 qdisc noqueue qlen 1000 > link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 > inet 127.0.0.1/8 scope host lo > valid_lft forever preferred_lft forever > inet6 ::1/128 scope host > valid_lft forever preferred_lft forever > 2: sit0(a)NONE: mtu 1480 qdisc noop qlen 1000 > link/sit 0.0.0.0 brd 0.0.0.0 > 3: mlan0: mtu 1500 qdisc mq qlen 10= 00 > link/ether d4:ca:6e:64:e3:61 brd ff:ff:ff:ff:ff:ff > inet 192.168.0.101/24 brd 192.168.0.255 scope global mlan0 > valid_lft forever preferred_lft forever > inet6 fe80::d6ca:6eff:fe64:e361/64 scope link > valid_lft forever preferred_lft forever > 4: wwan0: mtu 1500 qdisc pfifo_= fast q0 > link/[65534] > inet 10.114.34.107/29 brd 10.114.34.111 scope global wwan0 > valid_lft forever preferred_lft forever > = > The LTE connection came up with IP address: > = > wwan0 Link encap:UNSPEC HWaddr 00-00-00-00-00-00-00-00-00-00-00-00-0= 0-00-0 > inet addr:10.114.34.107 P-t-P:10.114.34.107 Mask:255.255.255= .248 > UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1500 Metric:1 > RX packets:253 errors:0 dropped:0 overruns:0 frame:0 > TX packets:348 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:1000 > RX bytes:28788 (28.1 KiB) TX bytes:37777 (36.8 KiB) > = > = > There is not much about the dump message: > = > # journalctl -u ofono > -- Logs begin at Fri 2019-07-05 05:07:43 UTC, end at Tue 2019-08-13 09:15= :10 UT- > Aug 11 12:07:12 systemd[1]: Starting Telephony service... > Aug 11 12:07:13 ofonod[189]: oFono version 1.24 > Aug 11 12:07:14 systemd[1]: Started Telephony service. > Aug 11 12:07:19 ofonod[189]: Interface org.ofono.AllowedAccessPoints not t > Aug 11 12:07:21 ofonod[189]: LTE attach IP type: 0 > Aug 12 06:30:19 ofonod[189]: LTE attach IP type: 0 > Aug 12 06:32:24 ofonod[189]: LTE attach IP type: 0 > Aug 12 06:33:28 ofonod[189]: LTE attach IP type: 0 > Aug 12 06:56:03 ofonod[189]: LTE attach IP type: 0 > Aug 12 06:56:06 ofonod[189]: LTE attach IP type: 0 > Aug 12 06:56:10 ofonod[189]: LTE attach IP type: 0 > Aug 13 08:49:09 ofonod[189]: Terminating > Aug 13 08:49:09 systemd[1]: Stopping Telephony service... > Aug 13 08:49:09 ofonod[189]: Exit > Aug 13 08:49:09 systemd[1]: Stopped Telephony service. > Aug 13 08:56:22 systemd[1]: Starting Telephony service... > Aug 13 08:56:22 ofonod[3315]: oFono version 1.24 > Aug 13 08:56:22 ofonod[3315]: Unable to hop onto D-Bus: Name already in ue > Aug 13 08:56:22 ofonod[3315]: Exit > Aug 13 08:56:22 systemd[1]: Started Telephony service. > Aug 13 08:57:29 systemd[1]: /lib/systemd/system/ofono.service:8: Executab" > Aug 13 09:00:30 systemd[1]: Starting Telephony service... > Aug 13 09:00:30 systemd[1]: Started Telephony service. > = >>>> # ofono/test/list-contexts >>>> [ /ubloxqmi_0 ] >>>> [ /ubloxqmi_0/context1 ] >>>> Name =3D Internet >>>> Type =3D internet >>>> AuthenticationMethod =3D chap >>>> Settings =3D { Method=3Dstatic Gateway=3D10.116.103.22 >>>> Netmask=3D255.255.255.25} >> >> No Address? No Interface? Running ofono with debug output will >> probably shed some light on what's going on here. > = > I think that could be because I only ran it once, if I ran > list-contexts 4 times consecutively, every time it displayed different > results included address and interface, is it the normal behaviour of > list-contexts? Sounds broken...??? > = > # /usr/lib/ofono/test/list-contexts > [ /ubloxqmi_0 ] > [ /ubloxqmi_0/context1 ] > AuthenticationMethod =3D chap > Active =3D 1 > Protocol =3D ip > Username =3D > Name =3D Internet > Type =3D internet > IPv6.Settings =3D { } > Password =3D > AccessPointName =3D telstra.m2m > Settings =3D { Gateway=3D10.114.34.108 Method=3Dstatic > Interface=3Dwwan0 Do UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1500 > Metric:1 > RX packets:253 errors:0 dropped:0 overruns:0 frame:0 > TX packets:348 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:1000 > RX bytes:28788 (28.1 KiB) TX bytes:37777 (36.8 KiB) > main} > = > # /usr/lib/ofono/test/list-contexts > [ /ubloxqmi_0 ] > [ /ubloxqmi_0/context1 ] > AccessPointName =3D telstra.m2m > Protocol =3D ip > IPv6.Settings =3D { } > Active =3D 1 > Name =3D Internet > Type =3D internet > Settings =3D { Method=3Dstatic Gateway=3D10.114.34.108 Netmask= =3D255.255.255.24} > Password =3D > AuthenticationMethod =3D chap > Username =3D > = > # /usr/lib/ofono/test/list-contexts > [ /ubloxqmi_0 ] > [ /ubloxqmi_0/context1 ] > Type =3D internet > AccessPointName =3D telstra.m2m > Settings =3D { Netmask=3D255.255.255.248 Interface=3Dwwan0 Addre= ss=3D10.114.34.} > Name =3D Internet > Username =3D > Password =3D > Protocol =3D ip > Active =3D 1 > IPv6.Settings =3D { } > AuthenticationMethod =3D chap > = >> static is correct. ofono configures the local network interface >> appropriately for the established context. >> >> Given that ofono isn't reporting either an address nor interface above, >> I wonder what the network interface has been configured to. What do 'ip >> link' and 'ip addr' show? > = > Since it is running in good status, debug may not make sense, but I > include here anyway: > = > # ip link > 1: lo: mtu 65536 qdisc noqueue qlen 1000 > link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 > 2: sit0(a)NONE: mtu 1480 qdisc noop qlen 1000 > link/sit 0.0.0.0 brd 0.0.0.0 > 3: mlan0: mtu 1500 qdisc mq qlen 10= 00 > link/ether d4:ca:6e:64:e3:61 brd ff:ff:ff:ff:ff:ff > 4: wwan0: mtu 1500 qdisc pfifo_= fast q0 > link/[65534] > = > # ip addr > 1: lo: mtu 65536 qdisc noqueue qlen 1000 > link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 > inet 127.0.0.1/8 scope host lo > valid_lft forever preferred_lft forever > inet6 ::1/128 scope host > valid_lft forever preferred_lft forever > 2: sit0(a)NONE: mtu 1480 qdisc noop qlen 1000 > link/sit 0.0.0.0 brd 0.0.0.0 > 3: mlan0: mtu 1500 qdisc mq qlen 10= 00 > link/ether d4:ca:6e:64:e3:61 brd ff:ff:ff:ff:ff:ff > inet 192.168.0.101/24 brd 192.168.0.255 scope global mlan0 > valid_lft forever preferred_lft forever > inet6 fe80::d6ca:6eff:fe64:e361/64 scope link > valid_lft forever preferred_lft forever > 4: wwan0: mtu 1500 qdisc pfifo_= fast q0 > link/[65534] > inet 10.114.34.107/29 brd 10.114.34.111 scope global wwan0 > valid_lft forever preferred_lft forever > = > I'll send you debug information when it is in bad connection status. > = > Thank you all for you kind help. > = > Kind regards, > = > - jupiter > = /Jonas --===============0071511886316583283==--