Ubuntu Networking :: Keep Getting Dropped From Network
Oct 11, 2010
my laptop running 10.10 and a dell. the college is witch i attend's unsecured wireless network keep disconnecting me from every on average 45 minutes to an hour ands will not re connect for about 10-15 minutes. it works fine anywhere else i go but here i get dropped.
I have been noticing a trend on boxes running 11.4 getting RX dropped packets where prior to 11.4 they did not get them. Basically every box running 11.4 now logs dropped RX packets. Have tried changing cables, and in one case replaced the NIC with a different brand (and driver) nic. Also played with receive buffers and yet the counters still increment. In addition to OpenSUSE, I run several operating systems to include Centos, Scientific Linux, Debian, OpenIndiana, HP-UX, VMS and so forth. So far this anomaly has only shown up on machines running OpenSUSE 11.4. Any one else noticing this on your OS 11.4 machines? If so, were you able to resolve?
I'm looking for an open source/free network emulator tool that I could use on Mac OS X, to simulate a slow network connection, limited bandwidth and other network characteristics such as dropped packets etc for both UDP/TCP connections (or even on the physical layer).
I'm looking for the simplest solution that would allow me to run TCP/UDP servers and have a few clients connect to them on localhost emulating various network connections. I'm mainly wondering if I can use something like Linux's netem on Mac OS X (or even better cross-platform Windows/Linux/Mac). Perhaps I can run VirtualBox and a Linux kernel running netem, has anyone had luck with that?[URL]...
My wireless connection keeps getting dropped very frequently. I am on a Dell XPS m1530 and I will just be on the interent and it just disconnects from wireless and I have to connect back. It is super annoying. I am on the latest version of ubuntu.
I've been using my current laptop since Ubuntu Edgy Eft and, since then, I never had any problem with wireless ( Network controller: Intel Corporation PRO/Wireless 3945ABG [Golan] Network Connection (rev 02) ) Since I upgraded to Lucid, my connection is dropped approximately every 3-4 minutes for 10s. Network-manager doesn't show anything but it is well visible with the netspeed applet that shows a warning sign "No connection" instead of the speed.This is very annoying and frustrating. I recall having seen the same issue in an blog article where the author was talking about his upgrade to lucid. It was just one line and I didn't care a lot at the time and cannot find the article back.
Is there anyone having heard about a possible issue ? (I've upgraded from Karmic when Lucid was still in Beta, so it might be the cause).
I'm working right now with ubuntu 11.04 on my asus ul30vt-x1, my wireless n network is working on a linksys e3000 router, my issue is basically with long file transfers. When the issue started i was reaching speeds of 100 Mbps through my wireless network even when I'm connected at 150 Mbps. However when at this speed, I don't know why, the connection is dropped after some minutes or even seconds. I made some testing on windows, I'm reaching 70 Mbps there and no problem at all with dropping the connection.
So my conclusion is that for some reason the linux driver, in my case, is not able to handle such speed of 100 Mbps. I know that a wireless network will have a throughput of aprox 60% the wireless speed. Right now im trying to apply traffic shapping to my transfers so i can get 70 to 80 Mbps, its being working so far.
One of our RHEL 5.3 servers has trouble about 30% of the time with TCP-based communications, but it does not seem to be firewall issues. From another computer on the same switch, you can SSH to the server sometimes and other times the SSH command will just hang. When it hangs, you can often just Ctrl+C and try it again and it works. Same with HTTP connections. You'll get part of a web page and then FireFox will just hang waiting for the rest and eventually time out. Same goes for communication initiated FROM the server. SSH'ing from the server to any outside server or connecting to any web site works sometimes, but most times not. iptables if off. No other firewalls are running. Tcpdump shows communication gets so far and then stops. It does not matter whether tou run tcpdump on that server or the client connecting to it. Either way you see the connection stops working. MEANWHILE, pinging with small or large packets works flawlessly. 10,000 packets, zero drops.
I've recently installed Ubunter 9.10 Server Edition to use as a NAT firewall for the lab I run. I'm using iptables to do NAT forwarding and everything works great except that, occasionally, connections seem to break. Ssh connections close with "Connection reset by peer" and HTTP connections just stall out.I believe this has to do with the firewall's internal network interface occasionally dropping packets.
with my other ethernet card problem solved, I suddenly run into this:
Code: eth1 Link encap:Ethernet HWaddr 00:02:e3:16:37:4c inet addr:10.0.2.1 Bcast:10.0.2.255 Mask:255.255.255.0 inet6 addr: fe80::202:e3ff:fe16:374c/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
This card was working perfectly fine up until....an hour ago and it started doing this. My iptables isn't blocking it somehow, because I didn't change anything. I tried reverting to an older kernel and that didn't help. It's not the network cable, it works fine in any other card. Also, the dropped packets seem to count down? It seems to go down by exactly one every time I run ifconfig, no matter the length of time in between running it.
when I change the hostname on my Ubuntu server, the DHCP clients table in my router is dropped.I'm running 32-bit Ubuntu Server 10.10. My router is a Linksys WRT54GL with firmware version 4.30.15. I'm changing the hostname by modifying /etc/hosts and /etc/hostname. I've also tried changing the hostname using the Linux hostname command in a terminal, but this does not seem to behave correctly according to the manpages (it does not actually change the hostname, or update either of the aforementioned files).
I've also tried using a different router - a Linksys BEFSR41 - but the DHCP table is still dropped when I change the hostname on my server (although this router is not all that different from my other one).
The strange thing is the connections coming through in the virtual machine seem to be getting dropped. A port scan from the internet says the port is closed, but it is not the case as I can see the connection coming through. As you've probably worked out, I am running a mail server. When I send myself test mail, the connections from the senders mail server also gets dropped.
I've run into a of a routing issue pertaining to packets leaving a firewall, traversing and IPSec tunnel, hitting the target and then returning via a different tunnel, finally arriving back on the source firewall but on a different interface from where it started. Once the packet has returned to the firewall it is dropped I've been unable to discover the reason for the drop. Two sides to the system, Firewall A and Firewall B. Each firewall provides the default gateway to its respective side and offers a backup IPSec tunnel to the high capacity tunnel handled internally. The Layer 3 Switch uses OSPF and takes care of the bulk of the behind the scenes routing between the sides. In case of failure the Layer 3 switches direct traffic to use the Firewall tunnels to route traffic.
I'm running a Debian Squeeze 6.0.1a box that's connected to my ISP via an L2TP connection that's managed by OpenL2TPD. The box is configured to perform NAT from local clients (on eth0) to the internet (on ppp0).
However, I'm having an issue with TCP packets that are sent from the box itself to the internet (packets originally coming from the local clients get sent and received over the internet just fine)
I'm using this Python app to test this:
Code: #!/usr/bin/env python import socket, time s = socket.socket(socket.AF_INET, socket.SOCK_STREAM) s.bind(('', 5003)) s.listen(1) while 1: conn, addr = s.accept()
I have a Netgear DG834 ADSL router (v3) which I have used for a couple of years. Prior to that I had a v2 of the same router, but upgraded for the one that supports ADSL 2+ (here in the UK).
Recently I have been having problems with the connection often being dropped, and when I access 192.168.0.1 for the router admin page, when it is not working I get no IP address listed for that of the ISP's server, and sometimes as well the upstream and downstream speeds are listed as 0 kbps.
Usually this was resolved by rebooting the router.
However, for the last couple of days, when using Ubuntu 9.04, the page of the router at 192.168.0.1 will not load. I have tried numerous different browsers, and none of them will give the router admin page.
When I tried booting from a live CD of Fedora 13, the router admin page came up just fine, no problems (although still not getting a proper connection to the Internet).
I was going to try to use the older router I had, but so far, as I cannot access the router admin page in Ubuntu, I have not done so (not had the time, and now I am at work posting this).
So there must be a problem in Ubuntu that is preventing it showing the router's admin page. The Network connection details show that the router has given the PC an IP address of 192.168.0.6 and the DNS servers are listed, so I know that Ubuntu is talking to the router, but it is not allowing me to access its settings in any way.
I'm running a ZTE MF112 HSUPA USB stick (3 broadband dongle) for wireless broadband using network-manager.I can connect fine to mobile broadband, but every now and then, the signal drops and the connection is disconnected. When I try to reconnect it goes straight to network disconnected. It looks like it's not even trying to connect.Then when I right click on network manager, disable mobile broadband, then enable it again, works fine until the next disconnect.The point is, why do I have to keep disabling / enabling it? Pulling the dongle out works too. But there should be a workaround.
I'm running ubuntu 11.04, and using firestarter for my firewall. Logging events, all good, but I'd like to be able to (tcp)dump the packets being dropped/rejected. Wireshark is great after the event, but can I hook iptables and tcpdump together somehow so the system logs the entire packet rather than the summary when it says "no"? I've had a quick look/google, but can't find anything. I'm sure somebody must have wanted to do this before - any pointers people?
PS - will keep looking and post the answer if I find one....
My wireless is flakey -- dropped connections, poor speed, etc. As diagnosis, I loaded the 'wavemon' utility so that I might watch the radio and signal and noise.Things will be good, then fall off, then get better, then go away entirely. Repeat continuously and randomly. For these experiments, my laptops -- more than one -- are on a table.The table is roughly 15-20 ft laterally from the access point. The access point is also 8 ft above the floor vs. 3 ft table height.(grin) While the signal "wobbles" everything else is mechanically stable and otherwise not moving.
As is known, there is a queue lies between the kernel subsystem and the network driver for incoming data. And if data come when this queue has no space for it, the data got dropped by kernel. Is there some way to see how many packets are dropped due to this buff penury? I tried netstat -s but could not find something useful. On the other hand, I found this 12176 packets collapsed in receive queue due to low socket buffer from netstat -s. I think this is something related to the per-socket buffer, but not the incoming queue between the network driver and kernel. Is this right?
Have Lucid installed on an Acer laptop. With an external keyboard plugged into a USB hub, then into the laptop, I have been losing the first character typed, sometimes. Most of the time it is OK, but it happens many times in an hour. If I am typing right along, no drops. Only the first character I enter is missing. Most of the time it is firefox I am typing into, but not always. My use of this machine is largely online, so probably that is coincidental.
This problem I have never seen before Lucid - I don't remember it in the beta ether, but I didn't run that too much.Versions 9.04 and before were OK (couldn't use 9.10 for other reasons).
That's probably everything pertinent. I have the SATA HDDs installed on SATA ports 1, 3, 4 (not the promise slots, 5 & 6), and the BD-ROM installed on SATA 2. I don't think that's an issue, though. I did replace a working sony opiartic (sp?) DVD-RW, so I don't suspect any cable issues (for that and for other reasons).
note: "restart" = "reboot via restart", as opposed to power down & turn back on.
1. I turn on the power. BIOS posts that it sees the BD-ROM in the SATA 2 spot.
2. I boot into Ubuntu 9.10 (updated with the latest x.20 kernel). It does take a little longer than usual to load between the grub menu and the login.
3. After login, the device shows up as cdrom0 in the "Places" menu. Clicking on it returns an error whether a disc is in the drive or not. At this point, however, the open/close button on the drive no longer works. Its as if it's lost all power or connection or something.
4. When I restart, BIOS hangs a while while posting. It shows only devices on SATA ports 1, 3, & 4. It's as if the BD-ROM doesn't exist anymore.
lspci and lspci -vvnn show nothing... well, technically they show everything BUT the drive. I'm planning on purchasing a second BD-ROM (different manufacturer, if I can help it), to test later, but I don't have high hopes. It could be a motherboard issue, I suppose, but updating the BIOS from 1.4 to the latest version (1.9) solved nothing. I don't suspect it's the hardware, though, since it works fine in Win7. I imagine that the culprit is linux/ubuntu, and I need some help figuring out if this is a fixable problem. I've been really, really happy with Ubuntu and linux in general. Ubuntu is my primary OS, with Win7 as my backup in case of emergencies, but I'm seriously considering wiping linux & using Win7 exclusively altogether with this severe hardware issue.
I was living peacefully with 10.04, till I upgraded to 10.10. The upgrade was smooth except for the ATI driver.Now I have no driver (nothing at System>Admin>Additional Drivers)glxgears does not exceed 60 fps. Life is too slow.Using synaptic, I installed fglrx, this broke everything!glxgears now gives segmentation fault. Booting is not graphical anymore, but rather a low-resolution `Ubuntu 10.10' big-font text.Can any1 help me install the correct ATI driver (if any) and get rid of the corrupted fglrx synaptic driver?
Script was still there in GreaseMonkey, with all it's pointers to the web presence. I have Google Chrome with the same extension and it stiill works fine. Thought maybe it might be an install problem, so uninstalled/reinstalled Greasemonkey and the script//no change. Uninstalled and reinstalled Firefox and extensions//no change.
I just dropped my laptop while it was on. It is a dual boot vista and Ubuntu machine (9.04) and Ubuntu was running at the time. The machine didn't go down and after I got done using it I shut it down normally. Now it will not come back up because grub is saying there is something wrong with the disk geometry. It did give me a command line, I forgot the exact error message but I can look it up if needed.
I tried plugging the hard drive into a different machine under a live cd but everything was in the lost and found folder (80 GB) and the folder was unreadable. I backed up the windows partition successfully using windows explorer and now have been looking at testdisk and Stellar Phoenix Linux Data Recovery ($80) for the Linux partition. I don't know if it is ext3 or ext4, I may have bumped it up to ext4 from the default ext3 during the install. Also a few files on the windows partition were unreadable, so the disk definitely has some physical damage. I'm getting a new one along with a back up system for sure after all this is over.
I got tired of figuring out testdisk so I let the paid app run its free test overnight to see if I could recovery anything before I bought it, but it got hung up on a sector halfway through.
Going forward I think I will try Photorec next and then testdisk again if that doesn't work. Does Photorec get almost all the normal office (doc xls), music (mp3) and video files (mp4) not just pictures?
I'm getting dropped frames from both Digital8 & MiniDV camcorders using Kino. MyCPU doesn't seem to be spiking. I've made sure compiz is disabled and no memory/cpu hungry apps are running. I'm using SATA drives so DMA shouldn't be the issue.
The video and audio that is captured seems to be fine.
We do NOT support samba on our Unbuntu servers but still zillions of windows machines are constantly trying to connect on the SMB ports. I've added a rule that drops access to destination ports 137-138 and that seems to work. But it creates many many log entries documenting that the packet has been dropped. I've been researching and cannot come up with a way to suppress logging for these drops.
I recently tried to upgrade my Slackware server's sendmail software from 8.14.3 to 8.14.4 and ran into a problem. In a nutshell all outbound non local mail is routed through a smart host requiring authentication. As a result I use a modified version of the "sendmail-slackware-tls-sasl-clamav.mc" to build the sendmail.cf. However, when I tried to run sendmail 8.14.4 with my newly created sendmail.cf I was greeted with the following.
Warning: Option: AuthMechanisms requires SASL support (-DSASL) Warning: Option: AuthOptions requires SASL support (-DSASL) Warning: Option: CACertPath requires TLS support Warning: Option: CACertFile requires TLS support
Apparently SSLv2 is now considered weak and the directive is a good idea, but at present this directive is not supported by the sendmail in this package.