PDA

View Full Version : [ubuntu] Really weird apt issue in server 9.10 i386



Tralce
January 13th, 2010, 06:03 PM
When installing through apt-get, or upgrading, downloads stop. I will be in the midst of downloading a file of any size (or even running apt-get update) and the transfer will simply stop. If I ctrl+c it, then start the install/update/upgrade over it will pick up where it left off, but it gets rather tedious as I have to do this 10-20 times during any upgrade. Occasionally, running /etc/init.d/networking restart will get me through the rest of a download uninterrupted.

This only happens on this specific computer (a PowerEdge 2650 with dual Gbit LAN cards) and none other on my network. I am using the same release on a Poweredge 1550 with no issues whatsoever.

Also, downloads with Firefox have no issues at all.

Any clus about what's going on here?

Tralce
February 7th, 2010, 01:58 AM
Please note that I have had no success in fixing this, and that it's extremely important that it gets sorted out!

mushwars
February 7th, 2010, 02:20 AM
this is a network issue not apt, are you using wireless, if so what driver/model card are you using, what wireless manager? do you experience temporary SLOW loading times on pages? Can you stream youtube videos without having to wait for them to buffer?

Tralce
February 7th, 2010, 02:45 AM
For some reason it obsessively drops pings from google.com and us.archive.ubuntu.com:

root@2uterror:~# ping us.archive.ubuntu.com
PING us.archive.ubuntu.com (91.189.88.45) 56(84) bytes of data.
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=1 ttl=46 time=132 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=2 ttl=46 time=138 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=3 ttl=46 time=144 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=54 ttl=46 time=151 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=55 ttl=46 time=159 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=56 ttl=46 time=142 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=57 ttl=46 time=156 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=58 ttl=46 time=171 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=59 ttl=46 time=234 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=60 ttl=46 time=240 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=61 ttl=46 time=148 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=110 ttl=46 time=194 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=111 ttl=46 time=153 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=112 ttl=46 time=159 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=113 ttl=46 time=148 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=114 ttl=46 time=195 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=115 ttl=46 time=151 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=116 ttl=46 time=153 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=117 ttl=46 time=151 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=118 ttl=46 time=137 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=167 ttl=46 time=199 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=168 ttl=46 time=170 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=170 ttl=46 time=217 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=171 ttl=46 time=134 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=172 ttl=46 time=137 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=173 ttl=46 time=156 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=174 ttl=46 time=143 ms
64 bytes from prat.canonical.com (91.189.88.45): icmp_seq=175 ttl=46 time=151 ms
^C
--- us.archive.ubuntu.com ping statistics ---
191 packets transmitted, 28 received, 85% packet loss, time 191321ms
rtt min/avg/max/mdev = 132.019/163.404/240.070/29.013 ms


So, now, the question is, "why is it dropping packets like that, and how do I go about fixing it?"

It's connected to the same switch as my desktop and two other laptops and they do not have this issue. It's a fresh install of Server 9.10 i386. The desktop has 9.10 x64 Desktop and the laptops have Desktop i386. Nothing else is different. Here's some info that may be pertinent.

root@2uterror:~# lspci|grep Ethernet
03:06.0 Ethernet controller: Broadcom Corporation NetXtreme BCM5703X Gigabit Ethernet (rev 02)
03:08.0 Ethernet controller: Broadcom Corporation NetXtreme BCM5703X Gigabit Ethernet (rev 02)
root@2uterror:~# ifconfig
eth0 Link encap:Ethernet HWaddr 00:0d:56:b9:c5:a1
inet addr:10.0.1.9 Bcast:10.0.1.255 Mask:255.255.255.0
inet6 addr: fe80::20d:56ff:feb9:c5a1/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:6803 errors:3 dropped:0 overruns:0 frame:6
TX packets:4559 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:808606 (808.6 KB) TX bytes:370583 (370.5 KB)
Interrupt:28

eth1 Link encap:Ethernet HWaddr 00:0d:56:b9:c5:a2
inet addr:10.0.1.10 Bcast:10.0.1.255 Mask:255.255.255.0
inet6 addr: fe80::20d:56ff:feb9:c5a2/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:8550 errors:0 dropped:0 overruns:0 frame:0
TX packets:6055 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:11689400 (11.6 MB) TX bytes:470343 (470.3 KB)
Interrupt:29