[SOLVED] Pacstrap operation too slow - Networking

osafi · 2015-11-13 14:23:20

Hi everyone!
I'm trying to set up Arch (for the first time, coming from Fedora) and I'm having trouble with pacstrap. First off, I'm using a wired connection with DHCP. As soon as I boot, ping -c 3 google.com responds fine

Running this command:

# pacstrap -i /mnt base base-devel

results in many time outs similar to:

error: failed retrieving file 'core.db' from arch.localmsp.org : Operation too slow. Less than 1 bytes/sec transferred the last 10 seconds

Things I've tried so far:
1. Disabled IPv6 using the ipv6.disable=1 kernel flag
2. Adjusted the mirrorlist
3. Booted the Antergos (same computer) installation image and ran 'pacman -Syy' and the database updated in no more than 5 seconds
4. Booted the Arch image on my laptop (wireless+dhcp) and ran 'pacman -Syy'. Again the database updated in a no more than 5 seconds, even with default mirrorlist

lspci -k | grep Eth -A 3

05:00.0 Ethernet controller: Marvell Technology Group Ltd. 88E8056 PCI-E Gigabit Ethernet Controller (rev 12)
        Subsystem: ASUSTeK Computer Inc. Motherboard
        Kernel driver in use: sky2
        Kernel modules: sky2
06:00.0 Ethernet controller: Marvell Technology Group Ltd. 88E8056 PCI-E Gigabit Ethernet Controller (rev 12)
        Subsystem: ASUSTeK Computer Inc. Motherboard
        Kernel driver in use: sky2
        Kernel modules: sky2

dmesg | grep sky2

[     2.785913] sky2: driver version 1.30
[     2.786009] sky2: 0000:06:00.0: Yukon-2 EC Ultra chip revision 3
[     2.786352] sky2: 0000:06:00.0 eth0: addr 48:5b:39:74:50:b5
[     2.786437] sky2: 0000:05:00.0: Yukon-2 EC Ultra chip revision 3
[     2.786693] sky2: 0000:05:00.0 eth1: addr 48:5b:39:74:44:70
[    36.150270] sky2: 0000:06:00.0 enp6s0: renamed from eth0
[    36.169918] sky2: 0000:05:00.0 enp5s0: renamed from eth1
[    38.053193] sky2: 0000:05:00.0 enp5s0: enabling interface
[    38.056459] sky2: 0000:06:00.0 enp6s0: enabling interface
[    39.867545] sky2: 0000:05:00.0 enp5s0: Link is up at 100 Mbps, full duplex, flow control both

ip addr

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536  qdisc noqueue state UNKNOWN mode DEFAULT group default
    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
2: enp6s0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc fq_codel state DOWN mode DEFAULT group default qlen 1000
    link/ether 48:5b:39:74:50:b5 brd ff:ff:ff:ff:ff:ff
3: enp5s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP mode DEFAULT group default qlen 1000
    link/ether 48:5b:39:74:44:70 brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.10/24 brd 192.168.1.255 scope global enp5s0
       valid_lft forever preferred_lft forever

ping -c 3 google.com

PING google.com (64.233.160.102) 56(84) bytes of data.
64 bytes from oi-in-f102.1e100.net (64.233.160.102): icmp_seq=1 ttl=47 time=53.8 ms
64 bytes from oi-in-f102.1e100.net (64.233.160.102): icmp_seq=2 ttl=47 time=58.2 ms
64 bytes from oi-in-f102.1e100.net (64.233.160.102): icmp_seq=3 ttl=47 time=38.6 ms

Does anybody have any idea what the problem could be or what to try next?
Thanks for any suggestions!

Last edited by osafi (2015-11-15 03:03:05)

TheChickenMan · 2015-11-13 15:17:00

Did you update your mirror list before running pacstrap?

# mv /etc/pacman.d/mirrorlist /etc/pacman.d/mirrorlist.bak
# rankmirrors /etc/pacman.d/mirrorlist.bak > /etc/pacman.d/mirrorlist

osafi · 2015-11-13 15:28:45

TheChickenMan wrote:

Did you update your mirror list before running pacstrap?
# mv /etc/pacman.d/mirrorlist /etc/pacman.d/mirrorlist.bak
# rankmirrors /etc/pacman.d/mirrorlist.bak > /etc/pacman.d/mirrorlist

Yup, I've tried configuring the mirrors (#2 on my list of tried things), though I didn't use that rankmirrors command. I just went to https://www.archlinux.org/mirrors/status/ and chose the top 3 US mirrors and moved them to the top of my mirrorlist file, the first being http://arch.localmsp.org/arch/

Also from #4 on my list, I was able to use the default mirrorlist (which has the French mirror first) on my laptop and it worked just fine so my connection to the mirror doesn't seem like the problem.

TheChickenMan · 2015-11-13 15:42:54

OK. I just wasn't sure what you meant by "adjusting" the mirrorlist and thought it was worth asking.
You could also try good bandwidth with your network connection. Beyond that I don't really have any better ideas right now.

$ curl http://speedtest.reliableservers.com/100MBtest.bin -o /dev/null

Last edited by TheChickenMan (2015-11-13 15:43:48)

osafi · 2015-11-13 23:10:27

TheChickenMan wrote:

OK. I just wasn't sure what you meant by "adjusting" the mirrorlist and thought it was worth asking.
You could also try good bandwidth with your network connection. Beyond that I don't really have any better ideas right now.
$ curl http://speedtest.reliableservers.com/100MBtest.bin -o /dev/null

Sorry, I should have made that clear, and thanks for the suggestion! Attempting to run that command, curl says it will take ~4 hours so there is some issue with my connection when booting the Arch image, but I don't know why.

On this same machine, running that command in my Fedora installation or the Antegos live cd takes ~15 seconds

TheChickenMan · 2015-11-14 00:53:59

Perhaps you could further narrow it down if you have the ability to host something on your local network. You could use a variation of the above command with curl to try and grab a file hosted on another computer on your LAN using http or ftp. This would prove that it's definitely a network problem and that it's definitely the computer's setup and not the router or switch doing something funny with your internet connection. Sorry but narrowing down the problem and hoping you run into a solution is the best I have. It looks like the network is configured correctly to me.

osafi · 2015-11-14 03:41:07

TheChickenMan wrote:

Perhaps you could further narrow it down if you have the ability to host something on your local network. You could use a variation of the above command with curl to try and grab a file hosted on another computer on your LAN using http or ftp. This would prove that it's definitely a network problem and that it's definitely the computer's setup and not the router or switch doing something funny with your internet connection. Sorry but narrowing down the problem and hoping you run into a solution is the best I have. It looks like the network is configured correctly to me.

Ok, I hosted a 100mb text file on my server and tried to download it with:

$ curl http://192.168.1.250:2080/100mb.txt -o /dev/null

It took ~1 hour for it to download it. I watched it more closely this time and noticed something odd about the 'Current Speed' number. It would frequently drop all the way down to 0 and stay on 0 for 10-20 seconds before starting back up again. I don't think it ever went longer than 20 seconds before it would drop back to 0 again. Any ideas on what could cause this to occur?

EDIT:
I still haven't figured it out, but it seems to be some issue with the way my network is configured (why it works for Windows/Fedora is beyond me).
I was able to install Arch using USB tethering with my phone, so I've gone ahead and marked the thread as solved. Thanks TheChickenMan for your suggestions!

Last edited by osafi (2015-11-15 03:09:05)

Arch Linux

#1 2015-11-13 14:23:20

[SOLVED] Pacstrap operation too slow - Networking

#2 2015-11-13 15:17:00

Re: [SOLVED] Pacstrap operation too slow - Networking

#3 2015-11-13 15:28:45

Re: [SOLVED] Pacstrap operation too slow - Networking

#4 2015-11-13 15:42:54

Re: [SOLVED] Pacstrap operation too slow - Networking

#5 2015-11-13 23:10:27

Re: [SOLVED] Pacstrap operation too slow - Networking

#6 2015-11-14 00:53:59

Re: [SOLVED] Pacstrap operation too slow - Networking

#7 2015-11-14 03:41:07

Re: [SOLVED] Pacstrap operation too slow - Networking

Board footer