Ubuntu Networking :: Entering Interactive Session SSH Hang
Apr 17, 2010
I am experiencing a 5-10 minute delay when trying to ssh into my ubuntu server.
root@localhost's password:
debug2: we sent a password packet, wait for reply
debug1: Authentication succeeded (password).
debug1: channel 0: new [client-session]
debug2: channel 0: send open
debug1: Requesting no-more-sessions@openssh.com
debug1: Entering interactive session. HANG
I've tried using the option UseDNS No in my my sshd_Config but to no success. I've tried reinstalling ubuntu fresh, but still it hangs upon trying to ssh from any machine. I am running Ubuntu 9.10 server.
Possible Duplicate:Forward SSH traffic through a middle machine.I am looking to get an interactive ssh session on a remote machine, but must login via a gateway.For example, right now I do the following:
From the output i am indeed able to login, but do not get an interactive session. I took inspiration for this attempt from using ssh to run a command remotely.
I am trying to write a perl script which will give an interactive session to a user to execute command on the server. I have written a small script to do this :
Code: !/usr/bin/perl -w use strict; use Net::SSH::Perl; my $host = '192.168.1.1'; my $username = 'user'; my $login_passwd = 'test123';
This problem has to do with Python's interactive help (To get to the interactive help, just type help() at any interactive session). I like to browse all my installed Python modules, to see if I can find something that would be useful for my hobby programming projects. However, recently, typing modules (which is the command to list all installed modules) freezes the interactive session.
Two machines with Lucid installed last week, both installed with the same standard settings. One machine can see the network and both itself and the other machine, the other PC cannot see the first machine or itself. In the Network window all I get is a folder saying 'Windows Network' and clicking that ticks away for a minute or so before saying error, cannot connect to it.
On both machines I have created a shared folder on the desktop to trigger an installation of samba shares which has happened.I cannot connect to the other PC by entering its IP address into location either.
I am complete new to the technical side of VOIP. I know above diagram is not technically correct. I want a setup that works like that and oh the cheaper yet not compromising the better, even ekiga or skype can do that.
Well, first off I'd like to say that I am much happier with the wireless on Ubuntu 10.04 than I was on 8.10 (the last time I tried Ubuntu). However, there is a small annoyance I'd like to fix: every time I log in, I get a prompt saying that an application (the wireless network manager) is requesting the password for such-and-such keyring. I have to enter the password before I can connect to my network. Is there a way to get around this somehow so that I don't have to enter the keyring password for this keyring (the "default" keyring) every time I log in?
I have a situation where I am in a non-interactive shell. I have tried from within my non-interative shell to spawn an interactive shell but my output still does not goto me. Isn't there a way I can somehow go into /proc or somwhere and make the output my /dev/tty1? Or some way else to remedy this?The situation arises because I drop from my restricted shell environment (a sort of CLI interface), into the actual Linux shell. I cannot change the code of the CLI environment I am just faced with being in the linux shell environment and its non-interactive. Its very annoying to have to put > /dev/tty1 after every command I type.
Not to mention it seems damn near impossible to get pagers like more and less to work properly when your in a non-interactive shell.
I just installed Ubuntu using WUI. It went OK. I am able to launch it, etc.
However I cannot manually enter my static IP information. I go to Edit Connections > IP4 settings > I then change the tab from DHCP to manual and then the "apply" button turns to a light darkish color and is non-selectable.
I also cannot save any network config note using gedit. It says I don't have permissions. I checked and changed my account type to Administrator and checked all the boxes then rebooted and still nothing.
All connections on my computer start loading and then hang: it happens when running apt-get update (a few repositories load successfully and then it hangs at a random one) and when loading web pages with the browser (it says "Connecting...", then "Waiting for....", sometimes some part of the page loads, and then it hangs).
This happens both with my wired and my wireless connection. It does not happen on Windows on the same computer.
My system hangs within a few minutes after I turn on the Intel Pro/Wireless 2915ABG interface built into my motherboard or if I boot with the interface turned on. I have been having this problem for a couple of years and tried to fix it myself from time to time. (I think that I even asked for help in the Forum once.) The wireless interface used to work fine with Ubuntu and I kept hoping that the problem would "magically" disappear with each upgrade. I also read about the recent problems with ndiswrapper and similar interfaces. But this doesn't seem to be my problem. The entries in /var/log/messages just prior to the last hang were
Dec 15 09:09:47 periwinkle rsyslogd: [origin software="rsyslogd" swVersion="4.2.0" x-pid="824" x-info="http://www.rsyslog.com"] rsyslogd was HUPed, type 'lightweight'. Dec 15 09:09:47 periwinkle rsyslogd: [origin software="rsyslogd" swVersion="4.2.0"
[code]....
I don't know what process gdb was trying to trace. Since I was testing the system, I was only running Firefox, which has always triggered the problem. The output from "dmesg | grep -e ndis -e ipw" was
In previous version of Ubuntu, the messages just before the hang used to complain about the firmware in the interface being buggy. I intended to send the output from iwconfig and "sudo iwlist eth1 scan" as well. But I can't turn on the interface without hanging the system and thereby destroying this post.
I just upgraded from the 64-bit version of Ubuntu 10.4 to 10.10 over the network using the Update Manager. But when I chose the "About Ubuntu" option from the System menu, I was surprised to see it say that I am running Natty Narwhal 11.4. My system is a Linux Certified LC2464T [URL]
School with linux running on students' laptops, connecting via wlan to a Debian NFS and LDAP server. Every student logs on his/her profile residing on the NFS server.The clients are set up with autofs. Earlier, I had set up the wireless network in /etc/network/interfaces, but this time I decided to configure network manager so as to bring up both wireless and wired network before logon. This setup has been working on for the last fire or five years with only minor changes. Also worked with Karmic Koala, but still with the interfaces file instead of networkmanager. The Vostro is also new here, we've previously used mostly Dell Latitude D505s.
So here is what works:
1: Clients can log on to LDAP and NFS servers both wired and wirelessly. Everything is smooth.
2: While on LAN, shutdown and restart works flawlessly (and quick as a breeze, I'm really impressed by startup/restart/shutdown times, under 25 secs!).
3: Shutdown and restart also works wirelessly when doing it either from a local account or from the GDM chooser.
What doesn't work, however, is shutting down or restarting directly from a networked account connected while only being connected over the wireless network. This is what's being displayed on the terminal after it has tried tho shut down for a while:
Code:
The system is going down for halt NOW!
acpid: exiting init: cron main process (1011) killed by TERM signal. init: tty1 main process (1365) killed by TERM signal.
[code]...
If I try ctrl-alt-del at this stage, it says:
"init: rc main process (3030) killed by TERM signal"
"Checking for running unattended-upgrades: "
And then it will hang again, until I hold the powerbutton for some seconds. The unattended-upgrades part is what seems to be the culprit. I suspect it is about the wireless network not being connected any longer or something like that, but I'm not sure about how to go about debugging shutdown scripts here. I'd be grateful for pointers. I will try and see how it goes with the old interfaces file setup, but I'd rather make nm work.
I have a small intel Atom based computer that acts as a 'combo' firewall / router / server / 802.11 Access Point that services my end user needs. I'm using hostapd 0.7.0 and Fedora 12 x86_64. The system works fine on Fedora 11. I decided to upgrade to Fedora 12 last night, and completely broke my wireless. Note that configurations are _identical_ between the Fedora 11 and Fedora 12 installation. Though, I did recompile the hostapd binary under Fedora 12 libs, etc.
Anyway, Client #1 associates fine. Client #2 associates, and hostapd hangs. Strace'ing the process shows nothing... just.. dead. SIGTERM works, but hostapd is unable to 'unallocate' the wlan0 interface, which appears to be in a hung state, so I can't simply re-start hostapd. So, I have to remove wlan0 from my bridge, them rmmod ath9k, modprobe ath9k, re-add wlan0 into the bridge, restart hostapd.. But the second association consistently hangs it every time. Next, I went and downloaded the latest compat-wireless drivers as of this posting. Selected the ath9k driver, built the modules, replaced them in my /lib/modules tree, depmod'd, reboot, etc..
BIG difference. Now hostapd does not hang, and there appears to be better logging/debug than ever before. This made me happy. Unfortunately, there appears to be a new problem. Clients can now associate, but only for a split second. They are immediately disconnected, and the following message appears via hostapd's stdout logging: hostapd: wlan0: STA 00:21:5c:39:a3:99 IEEE 802.11: did not acknowledge authentication response
For every client that attempts connection. Note that I'm using WPA2-EAP-PEAP w/ MSCHAPv2. I've also, for kicks, just tried regular WPA2-PSK with similar results. I'm using a mixture of clients, Windows XP, Windows 7, Linux Ubuntu, Fedora, and others.
Ive seen this a hundred times while searching google but I can't seem to get any of the fixes suggested to work for me.
Here are the specs code...
I have other servers on the same network with the same software/hardware that never loses its mount to the windows share, and nightly backups are run through those mounts. So why does the mount on this machine fail when we do a push? The only conclusion that I can come to is there is some sort of time out on the windows server that causes this. The other servers that have this same type mount use their share every night, where as the share on this server gets used once or twice a week. Once the mount hangs I can not unmount it, I have to reboot the server. Once the server is rebooted the push works fine. But then the next week when a push is tried it hangs. What else can I check?
UPDATE: I've also tried NFS mounts and autofs mounts and they hang as well
I enabled logging by echo 1 > cifsFYI and this is what I see in var/log/message code...
Ive removed the actual file names for security concerns. Any one have ideas as to why this is happening? The only other thing I can think of to try is to swap out the NIC but it's at a remote location so I can't do that right now.
Is there a session manager I can use with 10.10? I would like to try Openbox but am not sure how to select it as a startup session. I would like to be able to choose between kde, gnome and openbox.
I have a very bad attempt at hashing the components of an tcp session to assign/locate the session in a hash table bucket. I am pretty sure that it has a very high collision rate and when there are a very large number of tcp sessions my application is having to search a long linked list to find the session within the bucket.
All the hashing functions I have found take a single string input where I need to input several integers and hash them into a single result. My guess is that any real hashing function is going to produce better results than what I am currently doing.
I have multiple ubuntu machines and I connect to one through an NFS share. I have done this for a few years without issue. However, since re-installing ubuntu and upgrading to 10.4 I have a problem with my system hanging when the remote shares are lost.
Basically, I can power down the machine downstairs, and my main machine then has a fit. I can not open any folders in ubuntu, nor can I shut down. If I try and shut down the system hangs, last time it hung for 8 hours before I had to kill the power.
These are the lines in my fstab
I don't know what I've done wrong, or how I can prevent this from hanging. I have googled the heck out of this as well and can't seem to find an answer either.
I am putting together some new systems for my customer and I'm having some trouble with a script that we use to back up files to a DVD R. The problem is that I can't write a 2nd session to the DVD unless I eject the disk and reload it. The drives are slimline type drives, Sony BD-5730S and Teac DV-W28S-V93, so they won't reload without human intervention. Opsys is CentOS 5.4 or RHEL 5.4. I've tried both AMD and Intel based mother boards. If i try this on Fedora 11 or 12 it works fine. This works on IDE attached drives but not a SATA attached drives. Fedora appears to use something called genisoimage instead of mkisofs. I can't get genisoimage to run on CentOS or RHEL.
Here's the code to setup the test files:
rm -f /tmp/BDtest/* mkdir /tmp/BDtest dd if=/dev/zero of=/tmp/BDtest/blank.iso bs=10M count=1 for NUM in {1..160}
I'm currently using putty on my laptop to control my headless server (both Ubuntu 8.04.) I'm running an rsync job that ran all night long and still going all day today so far. I need to restart my laptop, but if I do, I'll break the SSH connection. If I SSH back in, I won't see the same thing (rsync is showing me each file it's copying) that I'm seeing now. (At least I think that's true. Please correct me if I'm wrong.)
So I guess my questions are..Is there some way to restore my SSH session when reconnecting? (So I can resume seeing which file is being copied and - more importantly - when the rsync job (finally) finishes?If not, how else can I know when the rsync job finishes?
I am very new to this and tried searching for this problem in the forums but have not find anything. When I installed ubuntu (formatted HD), the installer gave an error but then seemed to proceed as normal and everything installed (from CD). However, when I boot up from Live CD all my wireless works fine (it detects networks, connects, etc.) When I boot up without the CD it says networking disabled and gives me no other options.
I am connecting from my WinXP laptop to my Ubuntu Server desktop using PUTTY. The laptop is using Wifi and the desktop is on a wire. I have a D-Link DI-524 wireless router. My session works fine for a while, and then freezes, sometimes after only a few seconds, and sometimes after many minutes. The same thing happens when I use VMWare Server Console to connect from the laptop to the desktop's virtual machines. Other connections between these machines under this setup (RDP, samba, SVN) do not experience problems. I usually do not experience problems establishing the SSH and VMWare connections.I have ended up using Windows Remote Desktop to connect to a Windows XP virtual machine on the desktop, and opening a PUTTY session on that machine. That's a nice workaround, but it makes it so that I have to have that VM running in order to have a reliable SSH session to the box
I am currently in a project to set up an LTSP server with 10 thin clients. I am using Ubuntu 9.10 (Karmic).
Installing server and booting clients are working fine. Now, according to the need, I have to restrict user session numbers and allow resuming previous user session.
I have achieved to do the first one, but still could not able to setup the second one. As per requirement, if some thin can have power failure, the same session should be restored back. I am confused here, if I need to focus on saving xsessions or saving gnome sessions. I am looking for a concrete solution as I am running out of time.
When I started using Firefox It always used to prompt me if I wanted to save the session to start next time when it closed. I accidentally clicked not to ask me to do this action again. And now I want that pop-up box back and I can't find a setting to get it back
I have UBUNTU server 10.04 LTS with 3 NIC "eth0" local and eth1,2 as internet connection and it acts as firewall, http proxy and samba file server ,I installed Zentyal panel manager for my server for easier management I did not configure any specific rule for my firewall but I have some problem with my clients who wants to connect to my server as gateway or as file server even my self experienced these problems too. these problems are as follow:
1. some time for a few minutes (maximum 10 minutes) my server block some of my clients to access it or internet but just for minutes but it is very annoying. 2. all of my clients those who login to an https servers or login to their mail or those who has some software like team viewer say that they are logging out from their session randomly I mean some of them logging out from their mail(yahoomail or googlemail ) or disconnecting from teamviewer connection or as I saw team viewer disconnecting for a few seconds and then comes back again. but I did not set any thing in my firewall or other services. this is my complete iptable rules:
I need to run a script that prompts the user for some input and takes action based on that input. It must run at the end of the login process, after the desktop is displayed.I tried putting it in ~/.bash_login, but it didn't run. Do I need to tell /etc/profile to run ~/.bash_login? I thought it looked for it & ran it if it was found.
A related question would be - Can I make it so the terminal session that is running the script be the only thing that the user can do? In other words, they can't click on anything on the desktop or in the panels and have it run until after they respond to the input request.
The interactive map on weather.com no longer works for me in Firefox. I'm just looking for some feedback from other users if they can see the map or not. I'm using Lucid with current updates.
Whenever I am prompted to enter the root password, typical example would be in 'Power Management' clicking the button 'Make Default', pressing ENTER or clicking AUTHENTICATE doesn't close the box. To get passed this I have to press ALT+F4 and it continues which I'm assuming still does authenticate.
When I use rar e to extract a rar file, the request for entry of the password always says password will not be echoed. For longer passwords, this is difficult, since a single mistyped key will result in an error.
Does anyone know how I can use the rar command to extract and have the password echoed (appear on the screen as I type it)?
Yesterday, I did a clean install of Lucid on a T42 thinkpad, with wireless card as identified by lshw below. I did the installation from within a live CD session, with wifi enabled.
Code: description: Wireless interface product: PRO/Wireless LAN 2100 3B Mini PCI Adapter vendor: Intel Corporation
On reboot into the fresh system and update I had the issues that others have seen with network manager not connecting, repeatedly asking for password and so on. Oddly enough, using a Netgear usb wifi dongle worked perfectly, so I thought there might be a driver issue. Updates did not help. The network manager icon would not appear in the notification area at all when attempting to connect to wifi in a cafe locally. Today, I did a fresh re-install from the same CD with the home wifi switched off and no internet connection on the laptop. No issues. Wifi is working perfectly at present. I had to make the home connection available to all users to stop the keyring password request (I use autologin). I'll be trying to connect at local cafes tomorrow (one has no security, one uses an ascii wep password, and a third has a WPA encryption)
Is it possible that the settings from the live session are carried over in some way and interfering with network manager in some way? (This all started because my minimal install (see signature) reached a state where it would not show the network manager icon when attempting to connect to wifis in cafes and colleges, which I need to do. I decided to reinstall to return to a known state. I may try a minimal installation and manual wifi connection from the command line next...)
