Score:0

Ubuntu 20.04 LTS MTR _gateway reporting nearly 100% packetloss, clients experience stutters and lag

fr flag

I've recently come into some hassle with my dedicated server. I've done as much debugging as I can myself, from running tracerts and pings with a script. Watching the throughput of the server, its appropriate for its utilization, roughly 20mb/s in/out. The issue is, sometimes the server appears to become unresponsive to the outside network, however, it itself stays online and functionally connected. This causes all my clients to drop connection from the various services I host for about 2 minutes, sometimes longer. It appears as a major drop in packets, however I am unsure where in the routing something may be incorrect. Attached is some diagnostic information

MTR

 Host                                Loss%   Snt   Last   Avg  Best  Wrst StDev
 1. _gateway                         30.4%    24  7889. 1220.   1.7 7889. 2323.
 2. core-access-switch1-v4001.jnb1.h  4.2%    24    0.8   2.3   0.8   7.6   2.1
 3. (waiting for reply)
 4. google-2.jinx.net.za              0.0%    24    9.4  18.3   7.2  29.4   5.8
 5. 172.253.65.181                    0.0%    23   24.2  21.9   9.5  37.8   6.7
 6. 172.253.66.3                      0.0%    23   21.8  19.4   3.0  30.7   6.7
 7. dns.google                        0.0%    23   12.2  16.8   7.7  29.7   5.3

                                       Packets               Pings
 Host                                Loss%   Snt   Last   Avg  Best  Wrst StDev
 1. x.x.x.x                          77.8%    10   31.6  38.5  31.6  45.4   9.8
 2. core-access-switch1-v4001.jnb1.h 55.6%    10   31.4  36.7  31.4  44.2   6.2
 3. (waiting for reply)
 4. google-2.jinx.net.za             70.0%    10   33.8  33.8  33.8  33.9   0.0
 5. 172.253.65.181                   75.0%     9   33.0  32.7  32.4  33.0   0.4
 6. 172.253.66.3                     62.5%     9   33.2  33.2  33.0  33.4   0.2
 7. 8.8.8.8                          66.7%     9   32.4  32.2  32.1  32.4   0.1

ip route show

default via 156.38.x.x dev eno1 onlink
156.38.x.x/xdev eno1 proto kernel scope link src 156.38.x.x

ip -s link

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    RX: bytes  packets  errors  dropped overrun mcast
    8177784    82154    0       0       0       0
    TX: bytes  packets  errors  dropped carrier collsns
    8177784    82154    0       0       0       0
2: eno1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000
    link/ether ac:1f:6b:00:5c:94 brd ff:ff:ff:ff:ff:ff
    RX: bytes  packets  errors  dropped overrun mcast
    765764160485 2555872035 0       0       0       43888
    TX: bytes  packets  errors  dropped carrier collsns
    674858531647 1733340153 0       0       0       0
3: eno2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
    link/ether ac:1f:6b:00:5c:95 brd ff:ff:ff:ff:ff:ff
    RX: bytes  packets  errors  dropped overrun mcast
    0          0        0       0       0       0
    TX: bytes  packets  errors  dropped carrier collsns
    0          0        0       0       0       0

netstat -rN

Kernel IP routing table
Destination     Gateway         Genmask         Flags   MSS Window  irtt Iface
0.0.0.0         156.38.x.x      0.0.0.0         UG        0 0          0 eno1
156.38.x.x      0.0.0.0         255.255.255.248 U         0 0          0 eno1
djdomi avatar
za flag
looks like a connection over Wi-Fi for me, use cable instead
Kythosyer avatar
fr flag
This is a dedicated host from a datacenter, not some local computer @djdomi
mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.