Ubuntu Networking :: HTTP POST File Uploads On PCs - Getting 500 Internal Server Error Message
Sep 23, 2010
We're having an issue with HTTP POST file uploads on our two Ubuntu PCs. For some reason, whenever one of our users attempts to submit a file in an HTML form, the request times out, usually with a 500 Internal Server Error message. This problem is not limited to one site, but occurs on all sites that use file uploads. Also, the problem does not appear to be with our network, as a Windows 7 PC on the same network can upload files to the same sites without any difficulties. The problem is not browser-specific; we have tested with Firefox, Epiphany, and Google Chrome and all produce the same results. The issue is relatively new, and was first observed within the last month; before this time, both machines had no problems uploading files.
Does anyone have ANY idea what could be causing this? I've tried a number of things, including rebooting the PCs, rebooting the network, disabling IPv6, etc. I'm not very experienced in Linux system administration, but I can use the terminal and am familiar with some terminal-based diagnostic tools, so if you need any additional info or want me to try something, please let me know! I've exhausted my own computer knowledge with regards to finding a solution to this problem.
I have a registered domain that resolves (via dyndns) xxx.mydomain.com to my external facing router -easy no issues there. Behind the router I have several machines (some VMs) running webservers, mail etc... What I want to be able to do is redirect the external traffic based on xxx to the relevant internal machine and serve the content back to the external world.
I have tried using a http rewrite of xxx.mydomain.com to the relevant machine an it works fine from within my network, however externally the re-direct fails as the master DNS servers have no record of internal DNS setup in my network (obviously).
So is there anything I can do to get xxx. recognized externally? I'm only just starting to get my head around how DNS, HTTP, TCP etc all hang together. Am I barking up the wrong tree with rewrite? Should I be looking at proxys?
i am forwarding HTTP request to a internal server, it is quiet successful but access logs donot show the ip of the external m/c. Rather it shows the ip of the machine on which i have enabled port forwarding.
I am trying to connect to the web interface found at [URL] using curl. This first requires login information to be entered at [URL], but I am having an issue with the login process. I am trying to submit the following form via POST:
Code: <form action="j_security_check" method="post" id="login_form" name="login_form"> <center> <table style="background: #cac1cf;FONT-SIZE: 12px;"> <tr> <td align="center" colspan="2">Please enter your username and password:</td> </tr> <tr> <td align="right">Username</td> <td> <input name="j_username" style="width: 250px" id="j_username" type="text"/> </td> </tr> <tr> <td align="right">Password</td> <td> <input style="width: 250px" name="j_password" id="j_password" type="password"/> </td> </tr> <tr> <td colspan="2" align="center"> <input value="Enter" name="enter" type="submit"/> <input value="Clear" name="Clear" type="reset"/> </td> </tr> </table> </center> </form> The command that I am using for this is the following:
Code: curl -c cookies -b cookies -L -d "j_username=user%40domain.com&j_password=pass" [URL] The command is properly formatted as far as I can tell. I tested it with another website using a similar authentication scheme using different POST variables specific to the form and it worked fine.
When I run the above command with the -v tag, it reveals this: Code: * Connected to lcl.uniroma1.it (151.100.4.74) port 80 (#0) > POST /sso/j_security_check HTTP/1.1 > User-Agent: curl/7.21.0 (i686-pc-linux-gnu) libcurl/7.21.0 OpenSSL/0.9.8o zlib/1.2.3.4 libidn/1.18 > Host: lcl.uniroma1.it > Accept: */* > Content-Length: 44 > Content-Type: application/x-www-form-urlencoded > } [data not shown] < HTTP/1.1 408 The time allowed for the login process has been exceeded. If you wish to continue you must either click back twice and re-click the link you requested or close and re-open your browser < Date: Sat, 29 Jan 2011 15:26:41 GMT < Server: Apache-Coyote/1.1 < Content-Type: text/html;charset=utf-8 < Content-Length: 1554 < Connection: close < { [data not shown] 103 1554 100 1554 0 52 5081 170 --:--:-- --:--:-- --:--:-- 10223* Closing connection #0
I cannot tell why the login timeout is expired when I try to do this, and my investigation toward this end has been fruitless. I saw a brief snippet on Google that vaguely suggested that the underscores in the domain name were at fault, but replacing these with their encoded counterparts did nothing to resolve the issue (that, and underscores should be fine when sent unencoded according to the standards). I have extensively perused the man pages and have come up with nothing to adequately explain this behavior. I also talked to a friend who has worked with curl in his line of work, but he mostly has experience in the context of PHP and has not dealt with this issue before. I am running GNU/Linux 2.6.35-22-generic-pae.
I ran across the above article, which described a DoS attack in which requests are sent very slowly to the Web server. I'm running lighttpd 1.4.28 on a Gentoo Linux server, and I'm wondering if there is anything I could do in preparation to defend against such an attack.
A bug report [url] seems to indicate that there was a patch in place already against this sort of attack, but I wanted to be sure that was the same thing and if there was anything else I needed to do.
I have a really weird (but consistent) problem with my Kubuntu 10.10 install: I cannot post some HTTP forms.
First off, this is a client PC problem. My squirrelmail on the server works fine. I just use squirrelmail 1.4.17 to troubleshoot the ubuntu desktop problem
I used an old (07.04) Ubuntu install which worked fine. Then I wiped the disk and installed Kubuntu 10.10 on the same hardware. Everything works but **some** HTTP post does not work (I can log in but not send mail or save draft). I noticed I cannot log in to Yahoo, for example.
My webhosting account can display the apache access_log. When I hit the <Send> button the POST request never arrives to the web server.
I use a router (Dlink DL-604) behind a DSL modem and ooma box. There is a Windows 7 PC and a Kubuntu PC connected to the router. I can use squirrelmail just fine from the Windows PC.
I tried several steps: - reinstalled Kubuntu - installed Firefox and Chromium (on top of reconq) - ran from a CD on my other (Windows 7) PC - installed Wireshark and compared the traffic (but was unable to pinpoint a problem)
The result was the same: the <Send> button just keeps waiting; the POST request never makes it to the web server.
This sounds (and is) scary and suspect. The fact that the "demo" Kubuntu install (from the CD on my other Windows PC) using the reconq exhibits the same problem on a totally different hardware leads me to believe this may be related to Kubuntu. For example, I had to type this very message on the Windows PC as I could not post it on the forum from my Kubuntu box.
I think what i need to do is update the certifcate for the apache2, but I'm not sure how to do this, where to put it, and then which of the thousand apache config lines needs to be changed
Using netcat, nc(1), craft a valid http/1.1 request for getting http headers (not the html file itself!) for the main index page of www dot aalto dot fi. What request method did you use? Which headers did you need to send to the server? What was the status code for the request? Which headers did the server return? Explain the purpose of each header.
nc -v www dot aalto dot fi 8080 HEAD / HTML/1.1 host: www dot aalto dot fi And it returns: 200 OK Content-Length: 858 Content-Type: text/html Last-Modified: Thu, 02 Sep 2010 12:46:01 GMT [Code]....
I really don't know what does it mean. Question 2: Using netcat, nc(1), start a bogus web server listening on the loopback interface port 8080. Verify with netstat(, that the server really is listening where it should be. Direct your browser to the bogus server and capture the User-Agent: header "Direct your browser to the bogus server and capture the User-Agent: header" I don't understand this question.
I have a web server in my kitchen with apache running on it. Since the upload speed is quite low due to my isp I would like to execute a bash script that uploads a file to another server through a website (which is htaccess protected) The idea in general: Someone with access to my website browses through a folder, copies a file path to an input form and presses "upload". Rather than executing a bash script directly I could have a cron job running in background that finds the path and then uploads the file to the other server I have userspace on and is accessible via sftp/ssh. The file would be than erased later after a couple of days or so. That person would be able to access the file with higher speed some time later without logging in via ssh and doing all that manually.
There is a centralized FTP server which are accessed by clients to develop some web development project and to achieve this the clients are using ftp service. The problem is when they are trying to copy any files and directories they are getting the following error: "Message reported from the file system: FTP copy not supported,use move instead" I've checked all the permissions and they are all OK. I don't know why this is happening. they can create and move files and folders but can't copy files and folders.
I am using the following software stack:Linux version 2.6.32-21-generic (buildd@yellow) (gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5) ) #32-Ubuntu SMP Fri Apr 16 08:09:38 UTC 2010 (Ubuntu 2.6.32-21.32-generic 2.6.32.11+drm33.2) (Kubuntu) cisco anyconnect vpn client 2.3.2016, Mozilla Firefox 3.6.8
The problem I have is that once I join my company vpn, I have full access to corporate services confluence, jira, servers, etc. However when I use firefox to try and resolve a jira and post a body of text the connection timesout.
If i use any other browser it works fine, if slow, if i transition workflows it works fine, and if i use windows and firefox with the same cisco client it works fine.
This appears to be a specific issue with Firefox. I have noticed that in general firefox is slower on ubuntu than on any other platform.
I installed fedora 11 and configured xdsl(pppoe). I then performed an upgrade and installed some apps. Internet still worked before the restart.After that even though i could activate the device, i cannot view websites. I get a server not found error message. I have tried using both firefox and konqueror. Internet works in windowsIs this due to the upgrade? If so which upgrade would have caused the error?How do i resolve it
There are a few web databases (also including my own php-based pdf manipulator), where I need to fill a html form, and upload file attachments.
About one year ago, these sites stopped to work correctly, when using Firefox (but they work from Internet Explorer). The problem concerns file upload. Other users here also experienced this, and no firefox update corrected the problem in the past year (I am using Firefox 3.6.9 now, and the problem is still there).
When debugging my pdf creator, I found that the attachment-type of any file upload made by firefox is "text/html", irrespectively of what is the type of the uploaded file. Whilst files uploaded by IE have the correct attachment-type.
I have a Samba server conecting to a windows XP machine that is in workgroup named CLOUDLESS. I have a share folder that resides in the server that can be accessed no problem by the XP client, but I got the following error message from the syslog:
Code: Apr 5 18:15:44 firefly nmbd[5656]: find_domain_master_name_query_fail: Apr 5 18:15:44 firefly nmbd[5656]: Unable to find the Domain Master Browser name CLOUDLESS<1b> for the workgroup CLOUDLESS. Apr 5 18:15:44 firefly nmbd[5656]: Unable to sync browse lists in this workgroup.
I would like to find out how I would use both curl and wget to sent an http post to get the hostnames of a few servers. I know am not even given any work of anything I have done, but the reason is that I am really lost, and I do not even know how to start it.
I have been able (with some hair pulling and gnashing of teeth) to print from outside the LAN to the printer connected via USB to the desktop computer at home, until the last upgrade. The laptop will forst say the printer may not be connected, but eventually (usually) after a while says the job was sent to the printer. The log files on the desktop - serving as the printer server have an error message in cups/error_log that I am not familiar with, and don't believe I have seen before. I am sure if I could decipher that error and figure out how to change the behavior to get rid of it, it would once again work. Printing from within the LAN is a no-brainer as it has been for the last several releases. The error that I get when trying to print outside the LAN (via the internet) is:
cupsdAcceptClient: 12 from <WAN IP>:631 (IPv4) Unable to encrypt connection from <WAN IP> - GnuTLS internal error. I am missing something as far as authentication from the laptop to the desktop. how to fix it.
I agree deleting .Xauthority files work, but thats not a good solution. I mean, everytime I have to run any X forwarding application I have to delete the .Xauthority file. Here is my scenarioThe applications that I want to run are on Linux Server [NFS] and the client machines [CIFS] that are having problem in locking theXauthority files are Macs which share the same domain as that of client solaris machine. i.e. The home directory of particular user on Solaris & the home directory of that user on Mac have same contents.When I ssh -X from Solaris to the server, everything works fine, no error messages.When I ssh -X from MACs to the server, I get the following warning messages.
Code: /usr/X/bin/xauth: error in locking authority file /home/user/.Xauthority /usr/X/bin/xauth is the path on the server. If I try to break the lock by sudo
It's been awhile since I posted anything which is a good sign my install has been working well and I have been able to handle most everything. However, I'm not able to handle this issue. I recently installed F11 and everything went well. But, when trying to see my other computers on the local network, I cannot. I receive this error message: Unable to mount location Failed to receive shared list from server. I understand the message as it is obvious, but do not know how to fix it.
I use web server apache use php fastcgi (Centos 5.2, apache 2.2.9 , php 5.2.5 and fcgi-2.2.4 ). I receive error : "The server encountered an internal error or misconfiguration and was unable to complete your request. contact the server administrator, no-reply@zz.vn and inform them of the time the error occurred, and anything you might have done that may have caused the error. I view log :
I have installed vnc-server on my CentOS 5.6 virtual machine. I can connect to it with a java web browser so it seems to be working. However, I get the following error message when I start, stop or restart the vnc-server process. Quote: Starting VNC server: 1:ken xauth: timeout in locking authority file /root/.xauthkk661q
I have a debian box running Apache2 and PHP5.2.6 lenny.
When a request is made via https, php displays the content fine. If the request is made over HTTP the file is offered for download, rather than displaying it.
I know its probably something trivial but I've never seen this issue.
The plot thickens, I can display PHP over HTTP in some directories but not others (which offer the file for download)?
i am trying to get postfix to relay through my isp's smtp server, but i am getting an error.Host or domain name not found. Name service error for name=mail.grandecom.net type=A: Host not found, try again...i can send mail inside my domain, but when i try to send an email to a test gmail address i find the above error in the log.
A church I've been working with has a CCTV system that has a web interface for viewing the camera feeds. We need to see the page from the outside, but it is just an HTTP page, no encryption. The box itself does not accept any sort of SSL encryption. How I can get this on the net in a secure way? At worst I could set up a remote desktop type solution, but I was really hopping I could use some apache magic and just re transmit the page to https and ssl encrypted.
I am new to Linux and I wanted to setup a dedicated game server on my notebook.
I've installed Ubuntu 9.04. I than created a directory named /home/phlex/kfserver. I downloaded the application binary from Steam. I changed the file permission with chmod +x hldsupdatetool.bin and now the file attributes are -rwxr-xr-x.
The problem is when I try to run the binary file. If I type ./hldsupdatetool.bin I get an error message "-bash: ./hldsupdatetool.bin: No such file or directory. It seems the OS can't find the file. I've downloaded the file several times thinking it may have been corrupt but it's not. I can delete the file, I can rename it and so the file is there. Sometimes Linux can be aggravating to use.
I am working on a calendar system for my own use, basiccaly what's doing is downloading my list of appoints from Goole calendar using GoogleCL and saving it to a text file on my computer. I am refreshing this info once every hour...
What I'd like to do is check to see if any appoints are within the next 48 hours (I know how to do this part) and if so post a message to the ubuntu notification system so it gives a little popup... How would I go about this?
I am trying to install opencv on ubuntu. I have been following the directions exactly as given in the readme, and all goes well until I enter sudo make install, and get this error:
CMake Error at cmake_install.cmake:61 (FILE): file Internal CMake error when trying to open file: /home/iochinome/workspace/trunk/opencv/release/install_manifest.txt for writing.
I installed Nagios on my Ubuntu 10.04 server using apt-get and when I accessed the web console, everything was OK. I made some changes to apache (creating some new virtual sites) and since then Nagios gives me a warning message for HTTP with the message, HTTP WARNING: HTTP/1.1 404 Not Found. The sites that I created are working perfectly. I noticed that the attemps are 4/4. Does this need to be reset or does Nagios automatically reset that once it detects the issue is resolved?
I'm running a webserver and i've uploaded serveral .txt files. I want them to be downloadable... For example if someone opens: [URL], to start downloading, not just to open in the browser.
just curious, this might not be possible, but is there any way with ajax to open an ftp connection, download a file and then turn around and POST it to a web server? the reason i ask is because i'm writing a script on a shared hosting plan that doesn't give me permission to directly talk outbound on the ftp port but i need to synchronize a file from a csv on an ftp server, this means i have to do the process manually, is there any way i can just click the button and let my browser do the work?
