Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Eth0 keeps dropping, but works with reset. [SOLVED]
View unanswered posts
View posts from last 24 hours

 
Reply to topic    Gentoo Forums Forum Index Networking & Security
View previous topic :: View next topic  
Author Message
ferrelas
Tux's lil' helper
Tux's lil' helper


Joined: 14 Jun 2008
Posts: 109

PostPosted: Thu Jan 24, 2013 12:05 pm    Post subject: Eth0 keeps dropping, but works with reset. [SOLVED] Reply with quote

Solution: Reinstall, though it was most likely related to the udev update, and me not having CONFIG_DEVTMPFS enabled in the kernel.


My network connection keeps dropping and I have to do:

Code:
fconfig eth0 down
ifconfig eth0 up
/etc/init.d/net.eth0 restart


to get it to work. Then it works for 3~20 mins, then it just goes down again.

It started happening around the time I updated udev to 197-r3 and replaced module-init-tools with kmod, so I figured it might have something to do with kmod, after waging war on portage for a little while I managed to get it downgraded again, but the problem persisted. My kernel is kinda old (3.2.12), so I tried updating it (to 3.6.11), but that didn't work well at all, with that version eth0 is not created at all.

I'm not 100% sure that it isn't my hardware, but I don't have acess to any other hardware for testing right now. Internet works fine on my roommates laptop, and when my networking works, so I know that internet is working at least. I have no idea what to do now, any suggestions.


Last edited by ferrelas on Mon Jan 28, 2013 6:09 am; edited 1 time in total
Back to top
View user's profile Send private message
alacheesu
Tux's lil' helper
Tux's lil' helper


Joined: 01 Jun 2005
Posts: 113

PostPosted: Fri Jan 25, 2013 12:27 am    Post subject: Reply with quote

Does dmesg tell you anything? I had a similar problem with a realtek card at some point and it kept spamming my log with "link up" and "link down". It was eventually fixed in a later kernel version. What's the name of your NIC?

I don't know if upgrading your kernel will fix this (probably not), but you should probably do it anyway given that it's pretty old. How do you compile your kernel? My guess is that your card is still supported and you just need to make sure the correct module is compiled.
Back to top
View user's profile Send private message
ferrelas
Tux's lil' helper
Tux's lil' helper


Joined: 14 Jun 2008
Posts: 109

PostPosted: Fri Jan 25, 2013 2:08 am    Post subject: Reply with quote

I took a look at dmesg, but I don't see anything that looks awfully relevant, at least it isn't spamming link up and down:

Code:
[62936.798879] bridge-eth0: disabling the bridge on dev down
[62936.798900] bridge-eth0: down
[62936.798922] bridge-eth0: detached
[62936.995937] userif-3: sent link down event.
[62936.995939] userif-3: sent link up event.
[62938.590140] r8169 0000:02:00.0: eth0: link down
[62938.590146] r8169 0000:02:00.0: eth0: link down
[62938.590249] /dev/vmnet: open called by PID 2749 (vmnet-bridge)
[62938.590257] /dev/vmnet: hub 0 does not exist, allocating memory.
[62938.590274] /dev/vmnet: port on hub 0 successfully opened
[62938.590283] bridge-eth0: up
[62938.591652] ADDRCONF(NETDEV_UP): eth0: link is not ready
[62938.593043] bridge-eth0: attached
[62938.593081] bridge-eth0: disabling the bridge
[62938.603025] bridge-eth0: down
[62938.603029] bridge-eth0: detached
[62938.790376] userif-3: sent link down event.
[62938.790378] userif-3: sent link up event.
[62940.679209] r8169 0000:02:00.0: eth0: link down
[62940.680399] ADDRCONF(NETDEV_UP): eth0: link is not ready
[62943.031896] r8169 0000:02:00.0: eth0: link up
[62943.033073] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[62943.033255] /dev/vmnet: open called by PID 2749 (vmnet-bridge)
[62943.033264] /dev/vmnet: hub 0 does not exist, allocating memory.
[62943.033276] /dev/vmnet: port on hub 0 successfully opened
[62943.033285] bridge-eth0: up
[62943.033289] bridge-eth0: attached
[62943.233503] userif-3: sent link down event.
[62943.233510] userif-3: sent link up event.
[62948.712292] userif-3: sent link down event.
[62948.712294] userif-3: sent link up event.
[62953.074012] eth0: no IPv6 routers present


My NIC as listed in lspci is: "Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168 PCI Express Gigabit Ethernet controller (rev 06)", I use the r8169 driver, which is compiled into the kernel, I also have 8139cp and 8139too in there as well for some reason. It shouldn't be an issue related to the kernel, since everything has worked just fine (with this very kernel) untill recently. I'm also getting it to work in intervals, and for me that hints at a working driver, but then again, I'm no expert.

Ok, did some checking now and I figured out why it doesn't create with the new kernel at least (probably), it turns out I checked the box just below the one for my driver by misstake when configuring, that was a silly misstake. Unless there something else I really should try, I'll probably just give the new kernel a shot tomorrow morning. ;)
Back to top
View user's profile Send private message
ferrelas
Tux's lil' helper
Tux's lil' helper


Joined: 14 Jun 2008
Posts: 109

PostPosted: Fri Jan 25, 2013 11:03 am    Post subject: Reply with quote

Ok, that did not work, eth0 was not created with the new kernel even with the right driver.
Back to top
View user's profile Send private message
alacheesu
Tux's lil' helper
Tux's lil' helper


Joined: 01 Jun 2005
Posts: 113

PostPosted: Sat Jan 26, 2013 12:23 am    Post subject: Reply with quote

Seems like you have the same card as me:

03:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168 PCI Express Gigabit Ethernet controller (rev 03)

That is the card that kept going down for me. I don't remember which kernel version fixed it, but I believe it was one of the later ones in the 3.6 series (I'm using 3.7.3 now). The kernel network modules I have enabled are "Generic media independent interface device support", "Realtek devices" and "Realtek 8169 gigabit ethernet support".

I didn't actually have any problems until I set up a backup solution which caused large amounts of data to be transferred at high speed. With more modest transfer rates the card worked perfectly.

If a new kernel doesn't work for you and you can't find anything useful in your logs then I'm afraid I don't have any other suggestions.
Back to top
View user's profile Send private message
gentoo_ram
Guru
Guru


Joined: 25 Oct 2007
Posts: 474
Location: San Diego, California USA

PostPosted: Sat Jan 26, 2013 5:15 am    Post subject: Reply with quote

I've got bad news. I have one of those network chipsets on my motherboard. Constant link up/down messages. Network would just die. Not stable at all. I don't know if the problem is in the hardware or driver. But either way, the result is very bad. I ended up buying a nice Intel chipset PCI-e NIC. Much more stable; excellent drivers. My recommendation is to get another network card with a different chipset and save yourself a bunch of aggravation. Go Intel, they actually care about their Linux drivers.
Back to top
View user's profile Send private message
ferrelas
Tux's lil' helper
Tux's lil' helper


Joined: 14 Jun 2008
Posts: 109

PostPosted: Mon Jan 28, 2013 6:07 am    Post subject: Reply with quote

Well, I ran out of patience, and did this the hard way. A complete reinstall solved the issue. ;) Seems the problem was that I didn't have CONFIG_DEVTMPFS enabled. I didn't get the news item about it in time, because I didn't was to sync and update again before I had this solved, silly. :P


EDIT: Never had any problems with this NIC before this, so I think it's good. :)
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Networking & Security All times are GMT
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum