I noticed recently while playing online that my internet connection drops, disconnecting me from the multiplayer session, for then returning by itself a few seconds after, this seem to happen every 20/30 min, but it's mostly random. This seem to affect all the devices on my network since I observed similar behavior on my smartphone. This made me deduce the problem was either with my modem/router or with the ISP.
I contacted my internet provider for assistance, they made me run speed tests periodically on a single machine while disconnecting all the others, but all of them where as expected. In the end they found no problem on their side, so the culprit is probably my modem or router.
I'm no expert on networking, but I decided to run a ping test overnight to see if I could spot one of these drops, and it seems I found 2 of them, this is what the output looks like (this was done through a raspberry pi running linux):
2020-02-25 03:50:56 | 64 bytes from lhr35s10-in-f14.1e100.net (216.58.206.46): icmp_seq=14101 ttl=54 time=9.45 ms
2020-02-25 03:50:57 | 64 bytes from lhr35s10-in-f14.1e100.net (216.58.206.46): icmp_seq=14102 ttl=54 time=8.70 ms
2020-02-25 03:50:59 | From 192.168.15.1 (192.168.15.1) icmp_seq=14104 Packet filtered
2020-02-25 03:51:03 | From 192.168.15.1 (192.168.15.1) icmp_seq=14108 Packet filtered
2020-02-25 03:51:04 | From 192.168.15.1 (192.168.15.1) icmp_seq=14109 Packet filtered
2020-02-25 03:51:05 | From 192.168.15.1 (192.168.15.1) icmp_seq=14110 Packet filtered
2020-02-25 03:51:06 | From 192.168.15.1 (192.168.15.1) icmp_seq=14111 Packet filtered
2020-02-25 03:51:07 | From 192.168.15.1 (192.168.15.1) icmp_seq=14112 Packet filtered
2020-02-25 03:51:08 | From 192.168.15.1 (192.168.15.1) icmp_seq=14113 Packet filtered
2020-02-25 03:51:10 | From 192.168.15.1 (192.168.15.1) icmp_seq=14115 Packet filtered
2020-02-25 03:51:14 | 64 bytes from lhr35s10-in-f14.1e100.net (216.58.206.46): icmp_seq=14119 ttl=54 time=11.7 ms
2020-02-25 03:51:15 | 64 bytes from lhr35s10-in-f14.1e100.net (216.58.206.46): icmp_seq=14120 ttl=54 time=9.08 ms
and:
2020-02-25 05:17:27 | 64 bytes from lhr35s10-in-f14.1e100.net (216.58.206.46): icmp_seq=19284 ttl=54 time=8.63 ms
2020-02-25 05:17:28 | 64 bytes from lhr35s10-in-f14.1e100.net (216.58.206.46): icmp_seq=19285 ttl=54 time=8.34 ms
2020-02-25 05:17:31 | From 192.168.15.1 (192.168.15.1) icmp_seq=19288 Packet filtered
2020-02-25 05:17:36 | From 192.168.15.1 (192.168.15.1) icmp_seq=19292 Packet filtered
2020-02-25 05:17:37 | From 192.168.15.1 (192.168.15.1) icmp_seq=19293 Packet filtered
2020-02-25 05:17:38 | From 192.168.15.1 (192.168.15.1) icmp_seq=19294 Packet filtered
2020-02-25 05:17:40 | From 192.168.15.1 (192.168.15.1) icmp_seq=19296 Packet filtered
2020-02-25 05:17:41 | From 192.168.15.1 (192.168.15.1) icmp_seq=19297 Packet filtered
2020-02-25 05:17:42 | From 192.168.15.1 (192.168.15.1) icmp_seq=19298 Packet filtered
2020-02-25 05:17:43 | From 192.168.15.1 (192.168.15.1) icmp_seq=19299 Packet filtered
2020-02-25 05:17:47 | 64 bytes from lhr35s10-in-f14.1e100.net (216.58.206.46): icmp_seq=19303 ttl=54 time=17.6 ms
2020-02-25 05:17:48 | 64 bytes from lhr35s10-in-f14.1e100.net (216.58.206.46): icmp_seq=19304 ttl=54 time=9.56 ms
192.168.15.1 is the local IP of the modem. I'm not sure what Packet filtered
means, looking online it appears to be something about a firewall blocking the packets. A maybe relevant thing is that it seems that, if these are indeed drops in connection, the drops happen more frequently when there is more traffic, as there are only these two drops in the span on 10 hours during the night.
I don't know how I should proceed from here, is there some other diagnostics I should run? What could be possible workarounds or solution to these problem? I'm using a Digicom 8E4566 REW303-T05 and a Cisco SPA122.
Your help is greatly appreciated.
EDIT:
It seems that the Packet filtered
lines could be some other issue (or maybe not), as I discovered that when connection is lost, the ping
command just stops printing lines. I wrote a script to see where holes greater than 2 seconds between consecutive pings happen, and while there are holes of 4/5 seconds in the two instances reported above, 2 new holes appeared, this time lasting a whopping 31 seconds:
2020-02-25 10:10:08 | 64 bytes from lhr35s10-in-f14.1e100.net (216.58.206.46): icmp_seq=36816 ttl=54 time=10.1 ms
2020-02-25 10:10:39 | 64 bytes from lhr35s10-in-f14.1e100.net (216.58.206.46): icmp_seq=36846 ttl=54 time=8.94 ms
and:
2020-02-25 10:11:18 | 64 bytes from lhr35s10-in-f14.1e100.net (216.58.206.46): icmp_seq=36885 ttl=54 time=8.88 ms
2020-02-25 10:11:49 | 64 bytes from lhr35s10-in-f14.1e100.net (216.58.206.46): icmp_seq=36915 ttl=54 time=8.66 ms
EDIT2:
I found online a script that triggers a traceroute anytime ping fails. Most ping fails seem to be false positives as the respective traceroutes didn't fail, however I have 2 that show that the packets stopped at a local gateway:
FAIL on Wed 26 Feb 14:13:07 CET 2020
traceroute to www.google.com (216.58.198.36), 30 hops max, 60 byte packets
1 192.168.0.1 (192.168.0.1) 2.285 ms 3.446 ms 3.502 ms
2 192.168.15.1 (192.168.15.1) 6.141 ms 6.298 ms 6.358 ms
3 192.168.1.1 (192.168.1.1) 6.665 ms 7.402 ms 7.522 ms
4 * * *
5 * * *
6 * * *
7 * * *
8 * * *
9 * * *
10 * * *
--
FAIL on Wed 26 Feb 14:13:48 CET 2020
traceroute to www.google.com (216.58.198.36), 30 hops max, 60 byte packets
1 192.168.0.1 (192.168.0.1) 1.975 ms 3.759 ms 3.815 ms
2 192.168.15.1 (192.168.15.1) 5.907 ms 6.094 ms 6.154 ms
3 192.168.1.1 (192.168.1.1) 6.997 ms 7.179 ms 7.743 ms
4 * * *
5 * * *
6 * * *
7 * * *
8 * * *
9 * * *
10 * * *
There seems to be a "secondary modem" somewhere I wasn't aware of with local ip 192.168.1.1
, that I believe have been installed by my ISP. Since I've done this test over wifi I'm going to perform the same test over a wired connection and also see if I can spot similar drops in regular use, so that I may confirm this once and for all. Then I'll probably contact my provider and explain what I found.
192.168.15.1
?192.168.15.1
is the IP of the Cisco modem, while that of the router is192.168.0.1
.